Do not read the whole cluster chain for file on opening.
This removes restriction for opening files which span
within a large amount of clusters (>65k). That happened because
FrLdrTempAlloc cannot allocate more than about 256 KiB of memory.
Adjacent clusters for file are now read using one disk driver call.
Extracted from the pci.ids database at https://pci-ids.ucw.cz/ from 2019-09-28 03:15:02 .
Maintained by Albert Pool, Martin Mares, Michal Vaner, and other volunteers from the PCI ID Project.
- Don't match device id outside current vendor id list.
- Don't match subsystem id outside current device id list.
- Refactor some magic constants.
Addendum to 894635b.
- Match subclass properly, don't match third subclass.
- Don't match subclass from next class, add ClassEnd boundary.
- Use class name if subclass name not available.
- Gracefully return "Unknown" if no class name found.
Add localization for the file description of both XML and XSL file formats.
Please check for copypasta errors!
CORE-16390
Co-Authored-By: Stanislav Motylkov <x86corez@gmail.com>
Most likely not correct, as it breaks comctl32:buttonv6 tests
https://reactos.org/testman/compare.php?ids=68146,68147
But greatly eliminates perceived flickering when hovering
with the mouse about buttons in unthemed mode.
We suffered from these regressions for years and nobody did
find a proper solution, therefore it's time to at least workaround them.
The patch did not expose any new real-world-issues after I retested
the whole rapps suite against it in 0.4.12RCs.
I committed the same workaround already into 0.4.12-RC-30-g
9513c9bf12
CORE-13278 was once introduced/unhidden by SVN 73806 ==
c607de714c
CORE-16093 was once introduced/unhidden by SVN 74149 ==
301c476074
Need to cross check these changes with WINE ComCtl32 before you PANIC
HACK!!!!
Not in order:
Piotr Caban : Remove unneeded RECT parameter from CBPaintText helper.
Don't invalidate ComboBox on CB_SETCURSEL message.
Don't invalidate ComboBox on LBN_SELCHANGE and
LBN_SELCANCEL.
Add more CB_SETCURSEL tests on ComboBox.
Redraw combo text field even if it's empty.
Let ComboBox edit control handle the redraw even if
CBF_NOREDRAW is set.
Don't do the painting if combobox is not visible in
CBPaintText.
Dmitry Timoshkov : Don't force a combobox repaint on WM_SIZE.
This breaks z-order based painting and causes side effects
for
applications that during the WM_PAINT processing reference
internal
data associated with a not fully initialized window.
Ref :
2d9e3236ea
Fabian Maurer : Set listbox popup height correctly and add tests.
Properly set dropdown height.
Now that user32 and comctl32 combo are separated,
this won't lead to a regression anymore
Ref :
3d0be0bad8
Info by Sebastian Lackner Ref:
f0fc034997
Nikolay Sivov : Remove some misleading TODOs and confusing comments.
The submenu of desktop popup menu was erroneously shown at first right click.
- Fix the hit test position that was modified by window style change.
CORE-16382
This fixes a strange toggle-back-effect in 'Zeckensacks Glide Wrapper 0.84c'
by importing Wine commit
9dbe886e3f
Thanks to patches author Akihiro Sagawa <sagawa.aki@gmail.com>
v6 comboboxes default to displaying up to 30entries and
give a bad user experience in LiveCD
where we usually have low vertical screen resolution.
Therefore limit the height of the expanded boxes to
bring us close to how they were drawn before 0.4.12-dev-882-g
e3e173ffaa
Adding CBS_NOINTEGRALHEIGHT gives similar result for both:
comctl32 v5 and v6 comboboxes.
Similar commit for bootcd CORE-16252
went into 0.4.13-dev-740-g
8b67edd8fa
- Use tempnam() and fopen() instead.
- Prevent the use of file names with a leading slash or backslash.
- Also prevent the use of file names with a trailing dot.
- Remove temporary files after use.
According to the Advanced RISC Computing Specification v1.2, for partitions,
StartingAddress and EndingAddress are the start and end positions of the
partition in terms of byte offsets from the start of the disk.
CurrentAddress is the current offset into (i.e. relative to) the partition.
Fix also the FAT filesystem in accordance.
- FIXME fix: Retrieve the size of the disk in number of sectors in DiskOpen().
- Add extra validity checks in the DiskSeek() functions.
- Explicitly call PcDisk* functions in machpc.c and pcdisk.c, and
XboxDisk* functions in machxbox.c (the code in these files is not
called cross-platform).
CORE-16216 CORE-16248
- Modify DiskReportError() to use an internal "display counter", so that
imbricated DiskReportError(FALSE) ... DiskReportError(TRUE) can behave
as expected.
- Merge LoadAndBootDrive(), LoadAndBootPartition(),
LoadAndBootBootSector() into one LoadAndBootDevice() function + its
helpers (in progress...).
- "FrldrBootPath" -> "FrLdrBootPath";
- Get rid of GetFreeLoaderVersionString() and use FrLdrVersionString
directly instead;
- Rephrase some comments;
- Use RTL_NUMBER_OF();
- Reduce indent level in LoadOperatingSystem() and EditOperatingSystemEntry().
Bug fix:
- Fix DissectArcPath() for "ramdisk(0)" case.
- Move DiskStopFloppyMotor() calls into the implementations of
Boot(New)LinuxKernel() and Reboot() HW functions, and the explanation
comments in ChainLoadBiosBootSectorCode().
- Remove unneeded DiskStopFloppyMotor() dummies in ARM and PPC code.
- Use more adequate bitmask value to be sent to floppy's Digital Output
Register for shutting down its motor (based on OSDev & our floppy
controller driver).
- Add optional arguments BootDrive and BootPartition to
ChainLoadBiosBootSectorCode() so as not to modify explicitly the
FrldrBootDrive and FrldrBootPartition variables, that should remain
purely internal.
- Implement ChainLoadBiosBootSectorCode() for x64.
- Get rid of the machine-specific DiskGetBootPath(), and instead do its
job only once in the machine-specific InitializeBootDevices() (or in
MachInit() for PPC).
Cache the result of this operation into the globally-accessible
FrldrBootPath buffer. This avoids the unneeded calls to (Mach)DiskGetBootPath()
we used to do before.
Also remove the separate distinction between the PC and XBOX versions
of this functionality.
- Move the PC-specific DiskIsDriveRemovable() and DiskGetBootPath()
as well as the disk-IO-error functionality, back into the
corresponding PC-arch files.
- Simplify IniFileInitialize(), getting rid of IniOpenIniFile().
Also move the exports where they are on Windows. This was already handled correctly by MSVC and still doesn't fix the order on GCC builds, but that is a different story and shall be addressed at a different time.
They were wrongly pointing to the original target once rewritten
instead of pointing to the proper target: the device.
This notably fixes opening the MountMgr device from user
mode (to perform IOCTL calls, for instance), and might
also fix various bugs dealing with global namespaces.
This might have some various effects in ReactOS~.
This fixes the bug when VidInitialize asks for different translated addresses
and VgaIsPresent always returns FALSE either because the machine does not have
VGA controller at all (e.g. 86Duino Zero), or it does have video hardware
that is not compatible with VGA standard (e.g. Original Xbox).
The problem investigated by Stanislav Motylkov. Suggested fix by Thomas Faber.
CORE-14625 CORE-16216 CORE-16222
Co-authored-by: Thomas Faber <thomas.faber@reactos.org>
CORE-16356
Therefore the ERRORLEVEL value will be correctly set in accordance on
return from the utility in cmd.exe, and conditional tests based on this
returned value will succeed as they should.
- Reserve the pipe NtControlPipe0 for the security process.
- Count regular service control pipes from 1.
- Use I_ScIsSecurityProcess to identify the security process.
- Services.exe uses the SECURITY_SERVICES_STARTED event to notify the security process that NtControlPipe0 is ready for use.
CORE-9023
FIXES:
======
- Fix parsing of the multiboot options string.
NOTE: They are not yet treated in a case-insensitive manner!
- Fix a bug in ArcOpen() so that it correctly skips the first path
separator (after the adapter-controller-peripheral ARC descriptors).
The path separator can be either a backslash or a slash (both are
allowed according to the specs); they were also already handled
correctly in other parts of the code.
- Fix DissectArcPath() so as to:
* **OPTIONALLY** (and not mandatorily!) return the path part that follows
the ARC adapter-controller-peripheral elements in the ARC path;
* make it correctly handle the (yes, optional!!) partition() part in the
ARC path, for the multi(x)disk(y)rdisk(z) cases.
ENHANCEMENTS:
=============
- Directly retrieve the default OS entry as we enumerate them and
build their list (i.e. merge the GetDefaultOperatingSystem() helper
within InitOperatingSystemList()).
- Directly use the opened 'FreeLoader' INI section via its ID in the
different functions that need it.
- Make the custom-boot and linux loaders honour the boot options they are
supposed to support (see FREELDR.INI documentation / template).
This includes the 'BootDrive' and 'BootPartition' (alternatively the ARC
'BootPath').
This also allows them to take into account the user-specified choices in the
FreeLdr custom-boot editors.
- Modify the FreeLdr custom-boot editors so as to correctly honour
the priorities of the boot options as specified in the FREELDR.INI
documentation / template.
- Use stack trick (union of structs) to reduce stack usage in the
FreeLdr custom-boot editors, because there are strings buffers that are
used in an alternate manner.
- Extract out from the editors the LoadOperatingSystem() calls, and
move it back into OptionMenuCustomBoot(), so that when this latter
function is called there is no risk of having a stack almost full.
- When building the ARC-compatible argument vector for the loaders, add
the mandatory "SystemPartition" path. This allows the loaders to NOT
call the machine-specific MachDiskGetBootPath() later on (this data is
indeed passed to them by the boot manager part of FreeLdr).
- Improve the FsOpenFile() helper so as to make it:
* return an adequate ARC_STATUS instead of a mere uninformative BOOLEAN;
* take open options, as well as a default path (optional) that would be
prepended to the file name in case the latter is a relative one.
- Make RamDiskLoadVirtualFile() return an actual descriptive ARC_STATUS
value, and make it take an optional default path (same usage as the one
in FsOpenFile() ).
+ Remove useless NTAPI .
- UiInitialize() and TuiTextToColor(), TuiTextToFillStyle(): load or
convert named settings into corresponding values using setting table and
a tight for-loop, instead of duplicating 10x the same parameter reading
logic.
- UiInitialize(): Open the "Display" INI section just once. Remove usage
of DisplayModeText[] buffer.
- UiShowMessageBoxesInSection() and UiShowMessageBoxesInArgv(): reduce
code indentation level.
ENHANCEMENTS for NT OS loader:
==============================
- Don't use MachDiskGetBootPath() but use instead the "SystemPartition"
value passed via the ARC argument vector by the boot manager
(+ validation checks). Use it as the "default path" when calling
FsOpenFile() or loading the ramdisk.
- Honour the FreeLdr-specific "Hal=" and "Kernel=" options by converting
them into NT standard "/HAL=" and "/KERNEL=" options in the boot
command line.
Note that if the latter ones are already present on the standard "Options="
option line, they would take precedence over those passed via the separate
"Hal=" and "Kernel=" FreeLdr-specific options.
Also add some documentation links to Geoff Chappell's website about
how the default HAL and KERNEL names are chosen depending on the
detected underlying platform on which the NT OS loader is running.
[KMTESTS:RTL] (CORE-14565)
+ Added buffer overrun tests.
+ New style header.
+ ISO C90 compatibility
[DDK] (CORE-9819): RtlUnicodeStringPrintf(Ex) and RtlUnicodeStringValidate(Ex) fixes plus formatting.
[DDK]: Cosmetic changes (NTSTATUS variable naming and tabs)
[DDK] First implementation of Unicode functions for NtStrSafe (CORE-9819)
+ This implementation does not modify the interface of previous functions.
+ Since we need length returned for the destination UNICODE_STRING.Length field:
- an intermediate "Len" postfixed function was added with optional
argument to return the _vsnwprintf character count.
- Wide Printf(Ex) functions already existent work now call new Rtlp
private worker functions.
+ For several UnicodeStringPrintf(Ex) tests, see work going on CORE-14565
NOTES:
+ MSDN says they won't insert NULL characters on formatted strings. But
seems this is incorrect at MSDN (interpret this as DONT ASSUME NULL-termination).
We return zero-terminated strings, and MS functions seems to do the same.
(XP DDK 7.1)
[DDK] First implementation of Unicode functions for NtStrSafe (CORE-9819)
+ This implementation does not modify the interface of previous functions.
+ Since we need length returned for the destination UNICODE_STRING.Length field:
- an intermediate "Len" postfixed function was added with optional
argument to return the _vsnwprintf character count.
- Wide Printf(Ex) functions already existent work now call new Rtlp
private worker functions.
+ For several UnicodeStringPrintf(Ex) tests, see work going on CORE-14565
NOTES:
+ MSDN says they won't insert NULL characters on formatted strings. But
seems this is incorrect at MSDN (interpret this as DONT ASSUME NULL-termination).
We return zero-terminated strings, and MS functions seems to do the same.
(XP DDK 7.1)
[DDK] Fixed bad NtStrSafe.h va_start call (CORE-9819)
Update on RtlUnicodeStringPrintfEx plus cosmetic changes.
[DDK] First implementation of Unicode functions for NtStrSafe (CORE-9819)
+ This implementation does not modify the interface of previous functions.
+ Since we need length returned for the destination UNICODE_STRING.Length field:
- an intermediate "Len" postfixed function was added with optional
argument to return the _vsnwprintf character count.
- Wide Printf(Ex) functions already existent work now call new Rtlp
private worker functions.
+ For several UnicodeStringPrintf(Ex) tests, see work going on CORE-14565
NOTES:
+ MSDN says they won't insert NULL characters on formatted strings. But
seems this is incorrect at MSDN (interpret this as DONT ASSUME NULL-termination).
We return zero-terminated strings, and MS functions seems to do the same.
(XP DDK 7.1)
[KMTESTS:RTL] Tests for new RtlUnicodePrintf(Ex) functions (CORE-14565)
+ 45 Tests working.
+ Lacks remaining non-Unicode functions in NtStrSafe.h for now.
+ Used for first tests on CORE-9819
[KMTESTS/RTL] Fixed test not taking care of null (CORE-14565)
[DDK] Fixed bad NtStrSafe.h va_start call (CORE-9819)
[KMTESTS:RTL] First Test.
Update on RtlUnicodeStringPrintfEx plus cosmetic changes.
[KMTESTS] Added NtStrSafe to test lists (CORE-14565)
First commit: Skeleton for test implementation + Addition to COMMON SOURCES in kmtests Cmake lists.
WIP.
+Implementation of RtlStringVPrintfWorkerLenW, RtlStringVPrintfExWorkerLenW to avoid changing existing public functions. This is required as existent functions did not return vsnprintf result, so we didn't have any return info to update UNICODE_STRING.Length.
+Additional implementation of RtlUnicodeStringValidate for checking purposes.
+Former RtlStringVPrintfWorker(ex) refactored to RtlInternalStringVPrintf(Ex)WorkerW.
+No ANSI functions affected as we didn't need them for Unicode printf's.
WIP: RtlUnicodeStringPrintf
This fixes the crashes in HvpGetCellMapped on Windows Server 2003 when booting from Freeloader, as mentioned in maharmstone/btrfs#16.
When the bootloader loads the system hive, it cleans the data pertaining to any volatile keys. The Windows bootloader does this by setting SubKeyCounts[Volatile] to 0. After boot, the kernel marks any cell where this is 0 but SubKeyLists[Volatile] isn't HCELL_NIL as dirty, meaning that the sanitized version will then get flushed to the disk.
Because Freeloader sets SubKeyLists[Volatile] to HCELL_NIL straightaway, Windows thinks the cell is clean, and can unload it without flushing. If it then reads it from the disk, it will crash in HvpGetCellMapped due to the stale volatile pointers.
If you break on nt!CmpInitializeSystemHive on Windows and "gu" to the let the function run, you'll see that DirtyVector of the HHIVE has only the first 8 bits set. If you run it using the official bootloader, it'll have a lot more than that.
Nowadays ReactOS includes support for htmlhelp.h and a working implementation of HHCTRL.OCX. Although current imported HHCTRL.OCX suffers of this issue:
https://bugs.winehq.org/show_bug.cgi?id=47379
See also CORE-15019
+ Revert 'intptr_t' to 'size_t'.
- Add "Verdana", "Verdana Bold", "Verdana Bold Italic" and "Verdana Italic" fonts by duplicating "Bitstream Vera Sans" fonts and renaming them.
- Delete some "Verdana" font substitutes.
- Add LICENSE file.
I made these font files by using FontForge. Their family name is "Verdana". Their display names are "Verdana", "Verdana Bold", "Verdana Bold Italic" and "Verdana Italic". Their font filenames are verdana.ttf, verdanab.ttf, verdanaz.ttf and verdanai.ttf. The Underline Positions are hacked by adding an integer value for workaround of FontForge's bug. CORE-16264
CORE-16274
SETUPLIB:
=========
- Find the system partition initially when we create the list of
partitions.
- Split the old CheckActiveSystemPartition() helper in two helpers:
FindSupportedSystemPartition() and SetActivePartition(). This allows
simplifying slightly the former one, and allows the user, in an
interactive situation, to decide whether the "supported system
partition" found can actually be used or not.
- Remove the "OriginalSystemPartition" hack in the PARTLIST structure.
- Add a note regarding the SystemPartition member in PARTLIST.
USETUP:
=======
- Use the introduced helpers from above. If the "system" partition we
are going to use, in case we install ReactOS on a fixed disk, is *NOT*
the same as the original one (e.g. because it is detected to be not
supported by ReactOS...), display an informative screen to the user
and let him confirm whether or not he wants to change the partition.
If we install on a fixed disk, try to find a supported system partition
on the system. Otherwise if we install on a removable disk, use the
install partition as the system partition instead.
This allows providing a fix for CORE-16274.
SETUPLIB:
=========
- Remove useless HiddenSectors member in PARTENTRY structure.
- InsertDiskRegion() helper returns a BOOLEAN success.
- CreateInsertBlankRegion() helper sets LogicalPartition.
- Simplify the InitializePartitionEntry() helper so that its PartEntry
parameter is clearly the one that is being initialized (i.e. converted
from a blank region to) an actual partition, and use the helper
CreateInsertBlankRegion(). The calculations for the StartSector and
SectorCount are exactly equivalent with the old version of this
function. Also make it return a BOOLEAN success instead.
+ Add some extra validation checks.
+ Adjust CreatePrimaryPartition(), CreateExtendedPartition() and
CreateLogicalPartition() in this regard.
- Better handling of "RAW"-mounted partitions: treat them as
"Unformatted" only if they are RAW *AND* their PartitionType is one of
those associated with FAT file-system. Otherwise we cannot decide
whether they are indeed unformatted or have an unknown file-system on
them, therefore treat them as the latter.
In this regard, the IsSupportedActivePartition() helper should not
look for FileSystem == RAW but instead only look whether the partition
is Unformatted.
This should help with situations similar to the one described in CORE-16274
where a partition with a genuine file-system but not recognized by
ReactOS (because we currently do not have the EXT2/3/4 filesystem
driver loaded during 1st-stage setup due to commit 5a650f6b) and
therefore mounted as RAW, was thought to be unformatted.
USETUP:
=======
- Use the "global" SystemPartition pointer: this is the "system"
partition we will actually use. It can be different from the actual
one of the computer, for example when we install ReactOS on a
removable disk. This allows also to simplify the code.
- Remove the single-used DestinationDriveLetter variable.
- Remove BuildInstallPaths() helper and use InitDestinationPaths()
directly instead.
- Always mention the disk where the partition being formatted is.
- Cleanup old code comments, add assertions here & there...
This fixes regression CORE-16289 where we could not longer
move the taskbar at a different than default location,
as WindowSnap would interfere.
Many Thanks to the patches author Doug Lyons.
Eventually the heuristic that we use here to identify the
taskbar via used window-styles could be improved later.
Theoretically possible that it bails out on some other windows.
The regression was introduced by 0.4.12-dev-373-g
7e396787ed
* [NTOS:INBV] Move typedefs to the only single file where they are used.
* [NTOS:INBV] Refactor code & resources
Remove garbage
* [NTOS:INBV] Reduce fade time and remove wait for animation.
- Use a separate function for detecting blacklisted PCI slots
- Blacklist PCI-to-PCI bridge to avoid stack overflow on real hardware Microsoft Xbox 1.3
CORE-16319 CORE-16216
* [REG] Add save and restore help strings
* [REG] Implement REG SAVE
* [REG] Assert the BACKUP privilege
With this change applied, reg.exe correctly saves a key on Windows 10.
* [REG] Use RegSaveKeyEx()
* [REG] Implement REG RESTORE
* [REG] Close the process token
* [REG] Remove duplicate string
* [REG] Add missing comma
* [REG] Copy new strings into translation files
* [REG] Update STRING_USAGE translations
"Should be submitted to wine" (tm)
In my testing on Win2k3, EN_CHANGE is generated in OnInitDialog, that is correct. The target is property sheet. The property sheet does unchange the page after WM_INITDIALOG generation. CORE-16280
This fixes null pointer call bug, which was revealed in 85d44fc. Spotted by Mark Jansen. CORE-16216
Co-authored-by: Mark Jansen <mark.jansen@reactos.org>
Rename RosAudioSrv to AudioSrv in audio service itself and in all system components which are related to this, same as it done in Win2k3.
It allows MS DxDiag to detect the system audio service correctly, so it becomes possible to run DirectSound test properly with MS dsound.dll, although it works with some minor errors and only in older VirtualBox versions, ~ up to 5.1.38 (and in other emulators as well).
CORE-16307
- Added SAL annotation
- Simplified local variables
- Replace counterintuitive `do {} while ()` loops with nice `for` loops
- Get rid of goto and its label, it can be simply avoided
- Create xbox.cmake based on legacy.cmake and pic.cmake
- Revive Xbox HAL files and add them to the build
- Fix boot with Xbox HAL, add missing parts from halinit_up.c
- Code style and debug logging improvements in pcibus.c
- Allow to choose Xbox HAL in usetup
CORE-16216
- Replace the existing header files in favour to a pre-compiled header instead. Furthermore, move all the function prototypes in the pre-compiled header.
- As per the prototype definitions, LoadDataFromRegistry() and SaveDataToRegistry() mustn't accept any number of parameters.
- Rename rsrc.rc to osk.rc (main resource file) as most of user-mode applications name have their main resource as the same application's name anyways.
- When querying and saving the data to registry, use only one value, greatly reducing the bloat of variables (as per the Hermes Belusca's request on #1729 PR).
- Update the header description reflecting the official ReactOS Coding style guidelines.
* Implement ShutdownProcessTree in endproc.c which recursively kills process trees
* Include tlhelp32.h in precomp.h
* Check if the child process can be shut down
1. Right Click the Desktop.
2. Choose "Properties" menu item.
3. "Properties for Display" dialog must be shown.
The parameters of ShellExecuteW were wrong. The execution parameters must be separated from the file parameter.
CORE-16299
- Add some forgotten, or move some misplaced ArcClose() calls so as
to avoid leakages in failure paths. (Thanks to Serge Gautherie for
having caught some of them.)
- Simplify some code; remove unneeded casts; use string-safe functions.
- Factor out disk & partition opening in LoadAndBootPartition() and
LoadAndBootDrive() into a LoadAndBootPartitionOrDrive() helper.
- Use ARC paths and ArcOpen()/ArcRead() instead of calling the
machine-specific BIOS-oriented DiskGetPartitionEntry() and
MachDiskReadLogicalSectors() functions to open the disk/partition
and read their boot sectors.
- Don't forget to close the opened boot sector file in LoadAndBootBootSector().
- Add assertions for DiskReadBufferSize in PcMemFinalizeMemoryMap()
and EnumerateHarddisks().
- x86/amd64 only:
* Add a DisableA20 helper for disabling the A20 line, before rebooting
back, or chain-load a boot sector, into 16-bit world.
Also pulse the output port of the keyboard controller to clear out
its state after having set the state of the A20 line.
* In addition to disabling the A20 line when rebooting or chain-loading
a boot sector, reset the video back to 80x25 text mode.
- Reset the cursor position back to the origin when initializing or
terminating the TUI.
There were menu drawing regression after 03adec8141.
This PR will fix it. CORE-16244
- We don't have to draw the frame manually because of WS_EX_DLGMODALFRAME extended style.
- We don't need the margins (MENU_TOP_MARGIN and MENU_BOTTOM_MARGIN) in the client area any more because the client area doesn't contain the frame width.
- Use GetSystemMetrics(SM_CXDLGFRAME) and GetSystemMetrics(SM_CYDLGFRAME) for frame width.
- Get rid of the FsCloseFile(), FsReadFile(), FsGetFileInformation(),
FsGetFileSize() and FsSetFilePointer() wrappers and use the ARC
functions directly instead. Make FsOpenFile() return an ARC file
descriptor ID of the correct type. Get rid of unused FS_* defines.
- Use TRACEs in the ***Mount() filesystem functions for diagnostics
purposes.
- Remove a leak in FatGetFatEntry(). Assign stuff via QuadPart where
possible in FatMount(). Remove an unused member in FAT_FILE_INFO.
- Reduce code indentation in BtrFsMount() and remove a leak there.
- Disable reading the "BootPath" parameter in the linux loader since
we don't use this parameter (yet??)
We need to initialize SAM before the security services get started, otherwise we will have a running SamSs service although SAM is not initialized at that time. This is no good.
- Add a function that creates a new control set and deletes unused control sets in case of a successful boot.
- Add a stub function that will soon switch to the last known good control set in case of an unsuccessful boot.
CORE-16269
- Add IntGdiAddFontResourceEx function that is extended from IntGdiAddFontResource, in order to add dwFlags parameter.
- Add IntLoadFontsInRegistry function that will load the fonts from registry info.
- If loading from registry failed, IntLoadSystemFonts will be called.
- Use IntLoadFontsInRegistry rather than IntLoadSystemFonts in the OS startup.
- Add NameFromCharSet function.
- Append " (CharSetName)" to registry value name if not TrueType.
With this change we can remove the workaround from the off-screen
menus dropping below their parent.
Many thanks to Joachim for his help debugging / testing this.
CORE-16297
- EditBox: Display the initial contents of the text buffer.
This allows modifying already existing text in the passed buffer.
- Menu:
* Make both MenuHeader and MenuFooter optional (but the latter is
more "optional" than the former...).
* Allow passing a user-provided "Context" structure to the key-press
filter callback, and pass also the index of the menu item that has
been selected.
- Minor formatting fixes.
Add "Microsoft Sans Serif" fonts. CORE-16264
- Add "Microsoft Sans Serif" font by duplicating "Liberation Sans" font and renaming it.
- Add LICENSE file.
I made these font files by using FontForge. Its family name is "Microsoft Sans Serif". Its display name is "Microsoft Sans Serif". Its font filename is micross.ttf. The Underline Position is hacked by adding an integer value for workaround of FontForge's bug.
CORE-9023
- During loading and initialization of the list of operating systems
available in freeldr.ini, convert any legacy operating system entry
encountered -- they are like those in NTLDR's boot.ini file, i.e.:
ArcOsLoadPartition="LoadIdentifier" /List /of /Options
into a new operating system INI entry, like those used by default in
FreeLoader. This allows us to avoid treating this corner-case later in
different parts of the code. Also, the "BootType" value is now
determined there, only once.
- Convert the OS loaders entry-points to ARC-compatible ones, following
the "Advanced RISC Computing Specification, Version 1.2" specification
https://www.netbsd.org/docs/Hardware/Machines/ARC/riscspec.pdf
- Introduce helpers for retrieving options values from the argument vector
in a simple way.
- Simplify LoadOperatingSystem(), since now the "BootType" value has
been determined once while loading the list of OSes (see above) and
is well-defined there. Use the BuildArgvForOsLoader() helper to build
the ARC-compatible argument vector from the corresponding INI settings
for the selected operating system entry, and use it when calling the
corresponding OS loader.
- In the OS loaders, since we can now directly read the settings from
the argument vector (instead of using INI settings), we can avoid
using a bunch of fixed-size string buffers, and avoid potentially
failing IniOpenSection() calls as well.
- Simplify code in the Linux loader (and the RemoveQuotes() function).
- Add UiShowMessageBoxesInArgv() that acts on the "MessageBox=" settings
passed through the argument vector (equivalent to
UiShowMessageBoxesInSection() ).
- Use string-safe functions where needed (copy/concatenation/printf on
fixed-size buffers).
CORE-16294
We had added "Arial Black" font, so a request for heavy weight font will be realized as it is. This PR will fix font weight (FW_*) calculation of default menu items.
Helvetica must be font substitute to Arial font.
- Change some Helvetica substitutes to "Arial" from "Liberation Sans". Not all Helvetica. Some languages are lacking support.
Replace "Georgia" fonts with renamed "Gelasio". CORE-16293
- Replace "Georgia" fonts with renamed "Gelasio" fonts.
- Replace LICENSE file.
I made these font files by using FontForge. Their family name is "Georgia". Their display names are "Georgia", "Georgia Bold", "Georgia Bold Italic" and "Georgia Italic". Their font filenames are georgia.ttf, georgiab.ttf, georgiaz.ttf and georgiai.ttf. The Underline Positions are hacked by adding an integer value for workaround of FontForge's bug.
Gelasio is available at: https://github.com/SorkinType/Gelasio
Add 'Palatino Linotype' fonts with renamed 'TeXGyrePagella'. CORE-16292
- Replace 'Palatino Linotype' fonts with renamed "TeXGyrePagella".
- Replace LICENSE file.
I made these font files by using FontForge. Their family name is "Palatino Linotype". Their display names are "Palatino Linotype", "Palatino Linotype Bold", "Palatino Linotype Bold Italic" and "Palatino Linotype Italic". Their font filenames are pala.ttf, palab.ttf, palabi.ttf and palai.ttf. The Underline Positions are hacked by adding an integer value for workaround of FontForge's bug.
- Initialize BootPath and BootOptions buffers when fallback behaviour is not taken.
- Correctly skip all the understood whitespace (space & tabs) and the
quotes before reading the boot options when using the alternative syntax:
[Operating Systems]
section_name = "ReactOS" /bootoptions
Fixes the minor regression introduced in 370e8564 (r43875).
Add "Lucida Console" fonts. CORE-16264
- Add "Lucida Console" font by duplicating "DejaVu Sans Mono" font and renaming it.
- Delete some "Lucida Console" font substitutes.
- Add LICENSE file.
I used FontForge to rename it. Its family name is "Lucida Console". Its display name is "Lucida Console". Its font filename is lucon.ttf. The Underline Positions are hacked by adding an integer value for workaround of FontForge's bug.
Add "Arial Black" font. CORE-16264
- Add "Arial Black" font that is based on "Archivo Black" font of Omnibus-Type.
- Add LICENSE file.
"Archivo Black" is similar to MS "Arial Black" in its metrics. I used FontForge to rename the font. The font filename is ariblk.ttf. The family name is "Arial Black". The display name is "Arial Black". The Underline Position is hacked for workaround of FontForge's bug.
Add "Trebuchet MS" fonts. CORE-16272
- Add "Trebuchet MS", "Trebuchet MS Bold", "Trebuchet MS Bold Italic" and "Trebuchet MS Italic" fonts by duplicating "Open Sans" fonts and renaming them.
- Delete some "Trebuchet MS" font substitutes.
- Add LICENSE file.
I made these font files by using FontForge. Their family name is "Trebuchet MS". Their display names are "Trebuchet MS", "Trebuchet MS Bold", "Trebuchet MS Bold Italic" and "Trebuchet MS Italic". Their font filenames are trebuc.ttf, trebucbd.ttf, trebucbi.ttf and trebucit.ttf. The Underline Positions are hacked by adding an integer value for workaround of FontForge's bug.
Add "Times New Roman" fonts. CORE-16272
- Add "Times New Roman" and "Times New Roman Italic" fonts by duplicating "Liberation Serif" fonts and renaming them.
- Delete some "Times New Roman" font substitutes.
- Add LICENSE file.
I made these font files by using FontForge. Their family name is "Times New Roman". Their display names are "Times New Roman", "Times New Roman Bold", "Times New Roman Bold Italic" and "Times New Roman Italic". Their font filenames are times.ttf, timesbd.ttf, timesbi.ttf and timesi.ttf. The Underline Positions are hacked by adding an integer value for workaround of FontForge's bug.
Add "Franklin Gothic Medium" fonts. CORE-16272
- Add "Franklin Gothic Medium" and "Franklin Gothic Medium Italic" fonts by duplicating Alexei Vanyashin's "Libre Franklin" fonts and renaming them.
- Delete some "Franklin Gothic Medium" font substitutes.
- Add LICENSE file.
I made these font files by using FontForge. Their family name is "Franklin Gothic Medium". Their display names are "Franklin Gothic Medium" and "Franklin Gothic Medium Italic". Their font filenames are framd.ttf and framdit.ttf. The Underline Positions are hacked by adding an integer value for workaround of FontForge's bug.
Alexei Vanyashin's "Libre Franklin" fonts are available at https://github.com/alexeiva/Libre-Franklin .
Add "Palatino Linotype" fonts. CORE-16272
- Add "Palatino Linotype", "Palatino Linotype Bold", "Palatino Linotype Bold Italic" and "Palatino Linotype Italic" fonts by duplicating "DejaVu Serif" fonts and renaming them.
- Delete some "Palatino Linotype" font substitutes.
- Add LICENSE file.
I made these font files by using FontForge. Their family name is "Palatino Linotype". Their display names are "Palatino Linotype", "Palatino Linotype Bold", "Palatino Linotype Bold Italic" and "Palatino Linotype Italic". Their font filenames are pala.ttf, palab.ttf, palabi.ttf and palai.ttf. The Underline Positions are hacked by adding an integer value for workaround of FontForge's bug.
Add "Geogia" fonts. CORE-16272
- Add "Geogia", "Geogia Bold", "Geogia Bold Italic" and "Geogia Italic" fonts by duplicating FreeSerif fonts and renaming them.
- Delete some "Geogia" font substitutes.
- Add LICENSE file.
I made these font files by using FontForge. Their family name is "Geogia". Their display names are "Geogia", "Geogia Bold", "Geogia Bold Italic" and "Geogia Italic". Their font filenames are georgia.ttf, georgiab.ttf, georgiaz.ttf and georgiai.ttf. The Underline Positions are hacked by adding an integer value for workaround of FontForge's bug.
Add "Courier New" fonts. CORE-16272
- Add "Courier New", "Courier New Bold", "Courier New Bold Italic" and "Courier New Italic" fonts by duplicating FreeMono fonts and renaming them.
- Delete some "Courier New" font substitutes.
- Add LICENSE file.
I made these font files by using FontForge. Their family name is "Courier New". Their display names are "Courier New", "Courier New Bold", "Courier New Bold Italic" and "Courier New Italic". Their font filenames are cour.ttf, courbd.ttf, courbi.ttf and couri.ttf. The Underline Positions are hacked by adding an integer value for workaround of FontForge's bug.
Add Arial fonts. CORE-16272
- Add "Arial", "Arial Bold", "Arial Bold Italic" and "Arial Italic" fonts by duplicating Liberation Sans fonts and renaming them.
- Disable some Arial font substitutes.
- Add LICENSE file.
I made these font files by using FontForge. Their family name is "Arial". Their display names are "Arial", "Arial Bold", "Arial Bold Italic" and "Arial Italic". Their font filenames are arial.ttf, arialbd.ttf, arialbi.ttf and ariali.ttf. The Underline Positions are hacked by adding an integer value for workaround of FontForge's bug.
- Add some cleanup code in failure code paths, instead of asserting.
- Move BasepNotifyCsrOfThread() helper to the only file where it is used.
- Don't use ERROR_DBGBREAK in failure paths but just DPRINT the error
message: we handle the failures properly.
- When creating the remote thread, sync its service tag with the parent
thread's one.
- kernel32!BaseCreateStack() is compatible with ntdll!RtlpCreateUserStack().
- When checking whether a stack guard page can be added, its size has to
be accounted for in the checking logic.
- We have to satisfy the PEB::MinimumStackCommit constraint.
- We cannot use PEB::GuaranteedStackBytes in BaseCreateStack() since it is
nowhere initialized (default is 0). It gets initialized to a non-zero
value when the user manually calls SetThreadStackGuarantee().
https://www.installsetupconfig.com/win32programming/windowsthreadsprocessapis7_6.html
- RtlpCreateUserStack(): Fix memory leak in failure case.
- RtlpFreeUserStack() doesn't need to return anything.
See also commit 1bc59379 (r59868).
CORE-11319
v6 comboboxes default to displaying up to 30entries and
give a bad user experience in 2nd stage
where we usually have low vertical screen resolution.
Therefore limit the height of the expanded boxes to
bring us close to how they were drawn before 0.4.12-dev-882-g
e3e173ffaa
Adding CBS_NOINTEGRALHEIGHT gives similar result for both:
comctl32 v5 and v6 comboboxes.
- Token handle duplication must be the last step because we cannot close the duplicated token handle if something fails.
- Call LsaApLogonTerminated(), delete the logon session and free the profile buffer if something fails.
Fix and improve font dumping. CORE-16245
- FontGDI->FontObj is not a pointer.
- "%s" does null check, so we remove redundant null checks.
- Add FaceName and StyleName dumping.
- DiskGetFileInformation() should return relative addresses -- relative to the beginning of the "device" (partition, or disk) in question.
- FatXSearchDirectoryBufferForFile() should assign file attributes.
- Minor code style improvements in FatOpenVolume().
CORE-16216 CORE-16248
Co-authored-by: Victor Perevertkin <victor@perevertkin.ru>
- This allows to detect and dynamically handle different partitioning schemes.
- Implemented detection of MBR, GPT, Xbox-BRFR, and partitionless disks.
- Currently only MBR and Xbox-BRFR partitions are handled and tested.
CORE-9841 CORE-15768 CORE-16216 CORE-16248
- Do not use hardcoded ordinals, add missing '-private'.
Except for MSXML3, OLE32 and RSAENH, see commit bff824b2 for more details.
- SHSVCS: The 6 first exports are nonamed with ordinals. Replace exported
commented stubs by actual exported stubs instead.
Co-Authored-By: Timo Kreuzer <timo.kreuzer@reactos.org>
addendum to 0.4.13-dev-707-g
0e4db883e5
Jim Tabor intended to structurally use flTextAlign over
lTextAlign internally within IntExtTextOutW().
He initially committed these changes in 0.4.12-dev-805-g
3377fe184e
But we unintentionally lost these changes via 0.4.13-dev-370-g
8f482af0a8
Many thanks to Doug Lyons who helped a lot with fixing CORE-16133.
Add Comic Sans MS font files. CORE-16237
Add "Comic Relief" fonts by Jeff Davis under SIL Open Font License in the name of "Comic Sans MS". We cannot choose Comic Sans MS currently but it's font engine's bug.
I want to visualize the message stream and verify the states of each message. ROSTESTS-328
- Add a testcase named MessageStateAnalyzer into user32_apitest.
- Also fix type casts in <windowsx.h>.
CORE-13211
user32!DrawFrameControl drawed invalid menu arrow when the width is shorter than the height.
- Use the squared rectangle.
- Draw it in the transparent background mode.
- Save the old background mode of HDC.
This PR will fix UltraISO's bug.
Renewal of Marlett.ttf font file. CORE-16226, CORE-14872
- Based on Ged Murphy's Marlett.ttf (See 527f2f9057/media/fonts/Marlett.ttf).
- Erase the centered triangles.
- Adjustment of glyphs.
- Clarify the copyright ownership.
- Tested with 3 test programs (RadioAndCheck, PixelPerfectGlyph and MarlettCompare).
CORE-16123
- NETID: Correct the call when using the "Computer Name Change" in ReactOS.
- WINLOGON: Update the volatile "Hostname" and "Domain" variables from
their non-volatile counterparts.
Co-authored-by: Hermès Bélusca-Maïto <hermes.belusca-maito@reactos.org>
CORE-16181
This allows preventing spamming the time servers if the registry entry
becomes corrupted or missing. Also, disallow a setting of less than
2 minutes for the same reasons, and instead use the fallback of 9 hours.
We assume pso->sizlBitmap.cy is positive. CORE-15995
Because pso->sizlBitmap.cy is certificated as a positive value after initialization in SURFACE_AllocSurface, so we can safely omit abs and labs (absolute) function calls for it.
- Use an auxiliary list where we add the new enumerated faces, for easy
roll-back in case a failure happens during font loading. Only at the
very end when font loading has been successfully done, the auxiliary
list is appended to the corresponding global or per-proceess font list.
- Perform actual cleanup in case of failures.
- Introduce CleanupFontEntryEx() helper to simplify cleanup in
IntGdiLoadFontsFromMemory() in failure paths; implement
CleanupFontEntry() around it.
Additions:
- Perform string buffer NULL check in IntGdiAddFontResource().
- Reorder memory unmapping and section object dereferencing in
IntGdiAddFontResource().
- Move DuplicateUnicodeString() upwards.
- Mark explicitly the Marlett font handling hacks.
Add raster font (*.fnt and *.fon files) support. CORE-16165
- Add IntGetCharSet() function to get the charset by index and/or the number of charsets.
- Make IntGdiLoadFontsFromMemory() a non-recursive function.
- IntGetOutlineTextMetrics() accepts raster fonts.
- Improve CharMap handling (especially TT_PLATFORM_APPLE_UNICODE).
- Write the raster font file info to the registry.
- Don't request font size for raster fonts in IntRequestFontSize() function.
Fixes GCC 8 warning:
sdk/lib/crt/misc/i10output.c:83:25: error: '%d' directive writing between 1 and 11 bytes into a region of size 6 [-Werror=format-overflow=]
sprintf(format, "%%.%dle", prec);
^~
sdk/lib/crt/misc/i10output.c:83:21: note: directive argument in the range [-2147483648, 2147483646]
sprintf(format, "%%.%dle", prec);
^~~~~~~~~
sdk/lib/crt/misc/i10output.c:83:5: note: 'sprintf' output between 6 and 16 bytes into a destination of size 8
sprintf(format, "%%.%dle", prec);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Fixes GCC 8 warning:
sdk/lib/fslib/vfatlib/check/boot.c:173:27: error: '%02x' directive writing between 2 and 8 bytes into a region of size between 0 and 15 [-Werror=format-overflow=]
sprintf(buf, "%s%u:%02x/%02x", first ? "" : ", ",
^~~~
sdk/lib/fslib/vfatlib/check/boot.c:173:16: note: directive argument in the range [0, 2147483647]
sprintf(buf, "%s%u:%02x/%02x", first ? "" : ", ",
^~~~~~~~~~~~~~~~
sdk/lib/fslib/vfatlib/check/boot.c:173:3: note: 'sprintf' output between 8 and 31 bytes into a destination of size 20
sprintf(buf, "%s%u:%02x/%02x", first ? "" : ", ",
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
(unsigned)(p - (uint8_t *) b), *p, *q);
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Fixes GCC 8 warning:
dll/win32/oleaut32/oleaut.c:876:39: error: '%s' directive writing up to 99 bytes into a region of size between 92 and 291 [-Werror=format-overflow=]
sprintf(typelibkey, "Typelib\\%s\\%s\\0\\win%u", tlguid, ver, sizeof(void *) == 8 ? 64 : 32);
^~ ~~~
dll/win32/oleaut32/oleaut.c:876:5: note: 'sprintf' output between 18 and 316 bytes into a destination of size 300
sprintf(typelibkey, "Typelib\\%s\\%s\\0\\win%u", tlguid, ver, sizeof(void *) == 8 ? 64 : 32);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Fixes GCC 8 warnings like:
boot/freeldr/freeldr/disk/scsiport.c:806:31: error: 'partition(0)' directive writing 12 bytes into a region of size between 1 and 64 [-Werror=format-overflow=]
sprintf(PartitionName, "%spartition(0)", ArcName);
^~~~~~~~~~~~
boot/freeldr/freeldr/disk/scsiport.c:806:5: note: 'sprintf' output between 13 and 76 bytes into a destination of size 64
sprintf(PartitionName, "%spartition(0)", ArcName);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Fixes GCC 8 warning:
sdk/include/crt/mingw32/intrin_x86.h:76:12: error: 'memmove' offset [21, 40] from the object at 'DirContext' is out of the bounds of referenced subobject 'Attrib' with type 'unsigned char' at offset 19 [-Werror=array-bounds]
return memmove(dest, source, num);
^~~~~~~~~~~~~~~~~~~~~~~~~~
TOKEN_SOURCE::SourceString is an 8 char non-null-terminated string. Copy it as such.
Fixes GCC 8 warning:
dll/win32/msgina/lsa.c:195:5: error: 'strncpy' output truncated before terminating nul copying 8 bytes from a string of the same length [-Werror=stringop-truncation]
strncpy(TokenSource.SourceName, "User32 ", sizeof(TokenSource.SourceName));
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
TOKEN_SOURCE::SourceString is an 8 char non-null-terminated string. Copy it as such.
Fixes GCC 8 warning:
dll/win32/advapi32/misc/logon.c:638:5: error: 'strncpy' output truncated before terminating nul copying 8 bytes from a string of the same length [-Werror=stringop-truncation]
strncpy(TokenSource.SourceName, "Advapi ", sizeof(TokenSource.SourceName));
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Fixes GCC 8 warning:
In file included from base/applications/msconfig_new/xmldomparser.hpp:26,
from base/applications/msconfig_new/toolspage.cpp:11:
sdk/include/psdk/msxml2.h:12705:20: error: ISO C++ forbids converting a string constant to 'BSTR' {aka 'wchar_t*'} [-Werror=write-strings]
BSTR uri = L"x") = 0;
^~~~
* [FREELDR] Abstract VGA BIOS specific code
WinLdrSetupSpecialDataPointers() uses INT 10h video interrupts, but they are not available on Xbox, so make them machine-specific.
CORE-16204 CORE-16210
* [FREELDR] Abstract getting extended BIOS data area
WinLdrSetupSpecialDataPointers() uses INT 15h AH=C1h to get extended BIOS data area, but it's not available on Xbox, so make it machine-specific.
CORE-16204 CORE-16210
Make floppy detection code machine-specific, because Xbox CMOS cannot be used to detect floppies. Based on a patch by Matt Borgerson.
CORE-16204 CORE-16207
Co-authored-by: Matt Borgerson <contact@mborgerson.com>
- Refactor and use PC memory map init functions
- This fixes filesystem detection and probably some other problems
Based on a patch by Matt Borgerson.
CORE-16204 CORE-16206
Co-authored-by: Matt Borgerson <contact@mborgerson.com>
By removing font substitution for 'Microsoft Sans Serif'.
Thanks to the patches author Katayama Hirofumi MZ.
This is a partial revert of the guilty rev
master guilty 0.4.13-dev-9-g
811faed421
RC guilty 0.4.12-RC-13-g
8d3def0529
VBox test result:
https://reactos.org/testman/compare.php?ids=68497,68499
The same fix has been applied before to 0.4.12-RC-44-g
ca0e00e8af
- Add the following macros into <windowsx.h> header: HANDLE_WM_COPYDATA, HANDLE_WM_DEVICECHANGE, HANDLE_WM_DISPLAYCHANGE, HANDLE_WM_HOTKEY, and HANDLE_WM_MOUSEWHEEL.
- Fix HANDLE_WM_CONTEXTMENU macro.
They don't need to have a zeroed-out palette now that we can
automatically reset it from within the code.
Co-authored-by: Yaroslav Kibysh <yanet.prod@gmail.com>
This allows setting the memory protection of the kernel's resource
section as will. MmMakeKernelResourceSectionWritable() is re-implemented
around this helper.
This also makes it necessary to fix a bug in the previous code:
ZwEnumerateKey will not account for space for a null terminator, so to
ensure we have space, we must allocate the additional WCHAR, but not
include it in the buffer size passed to the function.
For MSVC, marking the section as discardable will do this automatically.
For GCC, we use a linker script that places it after the .reloc section
(which should be the last "real" section, check ld --verbose output for the
default linker script).
This fixes what seems to be a regression from r55835 (!).
Fixes a regression from 99fa38809f, which
replaced the kcom.h/stdunk.h versions (which zero memory) with local
implementations (which don't).
Standard C++ does not guarantee that memory is zeroed but several classes
rely on this, in particular for their m_Ref and m_bInitialized members.
Ideally the constructors should be fixed to initialize all required members,
but that task is error-prone so for now we simply restore the previous
behavior.
This allows users to have their own locally mapped DOS devices.
For instance, when a network drive is mapped, it is now mapped in the
user device maps and no longer in the global device map.
This avoids sharing mount points system wide whereas they shouldn't.
LUID mapped devices is the default W2K3 behavior.
Globally mapped devices can be restored by adding back this registry key.
https://twitter.com/HeisSpiter/status/1143615176450686976
CORE-16114
This allows getting rid of the ?? hack in the kernel but this doesn't
allow enabling LUID device maps as ReactOS can no longer open a
session with them enabled. So, we must remain with device maps at
root
CORE-16114
Patch :
Vijay Kiran Kamuju : Add DlgDirList wildcard checks.
Nikolay Sivov : Fix setting negative dialog item ids.
Alistair Leslie-Hughes : Don't reset focus if current dialog is a child.
Dmitry Timoshkov : If there is no dialog controls to set focus to then
set focus to dialog itself.
Zhiyi Zhang : Send notification for the focused button in
IsDialogMessage().
Huw Davies : Avoid using the comma operator in a while condition.
Patch skipped : a525631920d74e9d797f38305b89e63aed1c5c41 : Scale dialog
base units based on DPI awareness.
- MiLocateKernelSections(): Fix the calculation of MiKernelResourceEndPte,
MmPoolCodeEnd and MmPteCodeEnd.
- MmMakeKernelResourceSectionWritable(): Fix PTE looping upper limit;
use MI_MAKE_HARDWARE_PTE_KERNEL to build the updated read-write PTE.
- Introduce the MmMakeKernelResourceSectionWritable() helper for
making the kernel resource memory section writable, and use it
in KeGetBugMessageText(). Indeed, this latter function patches
in place the bugcheck resource message to trim any trailing
newlines before displaying the message on screen.
See also https://github.com/osresearch/uxen/blob/83bad53/dm/introspection-win7.c#L286
that mentions it too.
This fixes bugcheck text display (e.g. the MANUALLY_INITIATED_CRASH one)
when using (at least) MSVC-built ReactOS, avoiding a Page-Fault
exception during the bugcheck.
- Cover KeGetBugMessageText() in SEH since we are accessing kernel
resources that could also be corrupted in bugcheck scenarii, and we
don't want to further bugcheck.
- Fix newline trimming loop.
- KiDoBugCheckCallbacks():
* Wrap the bugcheck CallbackRoutine call in SEH.
* Add a FIXME concerning the need of further memory validation of CurrentRecord.
- Add a FIXME concerning the need to run the bugcheck-reason callbacks
with the KbCallbackReserved1 reason, in KeBugCheckWithTf().
Mentioned in http://blog.ptsecurity.com/2012/06/customizing-blue-screen-of-death.html
CORE-16116
- This allows getting the /NOGUIBOOT mode working correctly, as the latter
expects text-mode. Then, usetup (via BLUE driver) will change to a
different text-mode, or, GUI boot will change (via BOOTVID) to VGA
640*480 16-color.
- Always hide the VGA cursor.
- Remove also the deprecated boolean "Setup" in the *PrepareForReactOS()
functions.
Co-authored-by: Hermès Bélusca-Maïto <hermes.belusca-maito@reactos.org>
- Import cmdcnst.h and vga.h headers from the 'vga_new' VGA Miniport Driver,
that contain definitions related to VGA registers as well as command-stream
functionality.
- Replace a bunch of hardcoded values by their corresponding defintions.
- Replace "Captain-Obvious" comments in VgaIsPresent() with actual
explanations from the corresponding function in 'vga_new'.
- Simplify the VgaInterpretCmdStream() function, based on the
corresponding one from 'vga_new'.
- Use concise comments in the 'AT_Initialization' command stream definition.
- Import the 'VGA_640x480' initialization command stream from 'vga_new'
and use it as the full VGA initialization stream whenever the HAL
does not handle the VGA display (HalResetDisplay() returning FALSE).
Otherwise we just use the 'AT_Initialization' command stream that
performs minimal initialization.
- Remove unused AT_Initialization and other declarations from ARM build.
- VgaIsPresent(): Re-select the memory mode register before playing
around with the sequencer flag. Otherwise the VGA may not be detected
correctly from time to time.
- VidInitialize(): Simplify the initialization loop, reset AddressSpace
to its "default" value of 1 when calling HalFindBusAddressTranslation()
on 'NullAddress'.
This fixes:
- CORE-15911 "Scrollbars do not dissapear when Maximizing and not needed"
- CORE-10617 "ListView corrupt scrollbar upon resizing the column-header"
and will appease regression:
- CORE-15429 "Uninitialized scrollbars in 'My Computer'"
By applying clientFix.patch from CORE-15911 contributed by JIRA-user 'I_Kill_Bugs'.
https://reactos.org/testman/compare.php?ids=66723,66729
Earlier I already committed this fix into 0.4.12-RC-9-g
5ac548544f
This callback works. Haven't commit the rest yet. Still thinking about
the data packets types and shapes. Using wordpad to test this. It works,
loads and unloads due to incomplete driver support and no PDEV. It
crashes in ComCtl32 while loading a text file. When this gets committed
without a solution, the code will be commented out.
- Un-hardcode the RPL_MASK value.
- s/KiUnexpectedInterrupt&Number/KiUnexpectedInterrupt&Vector/
- Use C-style comments in traphdlr.c as everywhere else in the file.
- Update the URLs for the MSDN "FPO" macro documentation.
In CNotifyToolbar::AddButton tbBtn.iBitmap (imagelist-index)
was inialised with 0. Tigthvnc adds the icon wihtout NIF_ICON flag.
So the wrong icon index was kept and later replaced.
Step 1: Find the TcpIp transport and enumerate the installed network adapters using INetCfg.
Next Step: Find out how the binding process works and bind the TcpIp transport to all enumerated network adapters. This will replace the blind binding process we are using right now.
- Zoom factor is unsigned.
- It is comprised between 1 and 9, other values are invalid.
- Check for value validity when reading the zoom string from the combo-list.
Proposed changes:
- Use the total extent (width) and font metrics (height) as values
to put in the lpSize variable instead of calling ScriptString_pSize()
in the LpkGetTextExtentExPoint() function.
- Use GetTextExtentExPointWPri in more cases as another effort to get the data.
- Trailing whitespace fixes.
Pressing Next-button in setup of "1C Enterprise (training version) 8.3.10.2252"
did not have any effect anymore. Setup could not be completed anymore.
The regression was introduced to ros by 0.4.8-dev-739-g
f8b992f2d3
and was confirmed being a Wine-regression in
https://bugs.winehq.org/show_bug.cgi?id=45445
Thanks to Fabian Maurer for communicating to Wine and thanks to
Thomas Faber for regenerating cond.tab.c from the fixed source file cond.y
using BISON.
I also committed the same fix to 0.4.12-RC-32-g152ee20.
- Add a hack to rename the exported functions and use the Internal_ prefix for them so that the IImageList can use them with minimal code changes.
- Add the correct implementation of the affected exported functions making them now use the IImageList of the passed HIMAGELIST.
This indirection is possible because an HIMAGELIST is also a IImageList*.
This indirection is also needed because we can end up using image lists created by either comctl32 v5 or comctl32 v6.
This is how windows manage to use an HIMAGELIST created by comctl32 v5 when it is passed to comctl32 v6.
What really happens in this scenario is that the exported functions of both versions end up using the interface
stored in the object without checking if this interface is the version provided in the same dll or in the other.
- Windows requires 16 bytes of response data.
- Add the PLUGPLAY_CONTROL_USER_RESPONSE_DATA type.
- Usetup and Umpnpmgr must fail if NtPlugPlayControl(PlugPlayControlUserResponse) does not return STATUS_SUCCESS.
So that they handle LUID mappings and process device maps.
Get rid of the ObpParseSymbolicLinkToIoDeviceObject helper and introduce a
new helper ObpProcessDosDeviceSymbolicLink that will do the same things
but also handle name creation/deletion as well as device map handling.
All this is based on previous code (hence the same comments :-)).
What's left to do now is to add support for device maps in ObpLookupObjectName
Now for each request SenseBuffer is correctly set
and 3 attempts are made to be sure the STALL state is cleared and
the error is in something else.
Remove the usage of UFI_INQUIRY_DATA structure in favor of "standard"
INQUIRYDATA structure from scsi.h
Based in Vadim Galyant's patches to usbstor
- In the exported SHCreateDefaultContextMenu() and
IDataObject_Constructor() functions (called amongst others by
the exported CIDLData_CreateFromIDArray() function).
- In the exported SHCreateShellFolderView() function.
- In CDefView::GetItemObject(), where data was written to *ppvOut before
ppvOut was being checked for NULL.
by reverting essential parts of "Font rotation at arbitrary angles"
which was once committed in several steps:
Initial commit (PR#1207)
0.4.12-dev-243-g
f757a13519
broke font rendering for GTK apps CORE-15623
Second commit (PR#1238)
0.4.12-dev-558-g
64987cf273
restored font rendering in GTK apps but introduced slowdown CORE-16020
Katayama agreed to the temporal revert and will try to come up later
with a performance optimized version.
We revert in 0.4.12-RC and 0.4.13-dev.
The revert will cause failure-increases in gdi32:TextTransform
https://reactos.org/testman/compare.php?ids=67898,67901
* [ADVAPI32] Simplify RtlCreateUnicodeStringFromAsciiz() return value check
RtlCreateUnicodeStringFromAsciiz() returns a BOOLEAN, not a BOOL.
No functional change.
Addendum to
CORE-14271
* [USER32] Simplify RtlCreateUnicodeStringFromAsciiz() return value check
RtlCreateUnicodeStringFromAsciiz() returns a BOOLEAN, not a UINT.
Also, add a FIXME.
No functional change.
Addendum to
CORE-14271
* [USER32] Simplify RegisterClipboardFormatA/W() a bit
No functional change.
* [UDFS] Simplify SeSinglePrivilegeCheck() return value check
No functional change.
Addendum to
CORE-14271
This avoids at startup of Instant Messenger "QIP 2005 8095"
a message-box with "OLE Error 8004001".
The issue is a subtask of CORE-11537
Many Thanks to the patches author
Andreas Maier <staubim@quantentunnel.de>
JIRA-nick: andy-123
Note you still need to install Gecko 2.40 and Samba 1.3 from rapps for QIP.
It is now able to set the newly created device map to any
process and will default to current process if none is provided.
It also sets system device map if no process is specified.
It also deferences existing device map in the process if needed.
Finaly, it will make the directory object permanant.
Found by some static code analysis tool reporting 'Condition always false'.
The fix is a selective back-port from the more recent version of pICE hardware.c
Revision: 1.6, Date: 2004/02/17 23:07:36
The most recent source of pICE atm can be found at
https://sourceforge.net/projects/pice/files/pICE%20source/build_20/
Thanks to DougLyons and JIRA user 'maddin200'.
Split win32knt_apitest to 3 modules (win32knt_apitest, win32knt_xpsp2_apitest and win32knt_2k3sp2_apitest) for usability. These 3 modules are different in linking (win32u.dll, win32u_xpsp2.dll and win32u_2k3sp2.dll). I think it's better to separate for clarity.
Fixes GCC 8 warning:
base/applications/network/telnet/src/tnmain.cpp:171:8: error: this 'for' clause does not guard... [-Werror=misleading-indentation]
for (j = cursor; j >= 0; j--)
^~~
base/applications/network/telnet/src/tnmain.cpp:174:9: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the 'for'
for (k = --j; k >= 0; k--)
^~~
Fixes GCC 8 warning:
base/applications/network/telnet/src/tmapldr.cpp:70:3: error: this 'while' clause does not guard... [-Werror=misleading-indentation]
while (buf[len])
^~~~~
base/applications/network/telnet/src/tmapldr.cpp:74:4: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the 'while'
if (len && (buf[len-1] == ' ')) {
^~
Fixes GCC 8 warning:
base/applications/network/telnet/src/tnmain.cpp:171:8: error: this 'for' clause does not guard... [-Werror=misleading-indentation]
for (j = cursor; j >= 0; j--)
^~~
base/applications/network/telnet/src/tnmain.cpp:174:9: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the 'for'
for (k = --j; k >= 0; k--)
^~~
Fixes GCC 8 warning:
dll/win32/browseui/explorerband.cpp:1332:17: error: this 'if' clause does not guard... [-Werror=misleading-indentation]
if (!SUCCEEDED(hr) || !pParent.p)
^~
dll/win32/browseui/explorerband.cpp:1335:21: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the 'if'
hr = pParent->SetNameOf(0, pidlChild, dispInfo->item.pszText, SHGDN_INFOLDER, &pidlNew);
^~
Fixes GCC 8 warnings:
dll/opengl/glu32/src/libnurbs/internals/subdivider.cc:852:2: error: this 'if' clause does not guard... [-Werror=misleading-indentation]
if( jarc->pwlArc ) jarc->pwlArc->deleteMe( pwlarcpool ); jarc->pwlArc = 0;
^~
dll/opengl/glu32/src/libnurbs/internals/subdivider.cc:852:59: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the 'if'
if( jarc->pwlArc ) jarc->pwlArc->deleteMe( pwlarcpool ); jarc->pwlArc = 0;
^~~~
Fixes GCC 8 warning:
dll/win32/mshtml/script.c:844:4: error: this 'if' clause does not guard... [-Werror=misleading-indentation]
if(!new_buf)
^~
dll/win32/mshtml/script.c:846:13: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the 'if'
This->size <<= 1;
^~~~
Fixes GCC 8 warning:
drivers/filesystems/reiserfs/src/memory.c:874:5: error: this 'if' clause does not guard... [-Werror=misleading-indentation]
if (Mcb->RfsdFcb)
^~
drivers/filesystems/reiserfs/src/memory.c:877:2: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the 'if'
if (RfsdLoadInode(Vcb, &(Mcb->Key), &RfsdIno)) {
^~
Fixes GCC 8 warning:
sdk/lib/3rdparty/zlib/contrib/minizip/zip.c:521:5: error: this 'for' clause does not guard... [-Werror=misleading-indentation]
for (i=(int)uReadSize-3; (i--)>0;)
^~~
sdk/lib/3rdparty/zlib/contrib/minizip/zip.c:529:7: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the 'for'
if (uPosFound!=0)
^~
Fixes GCC 8 warning:
dll/3rdparty/libxslt/security.c:358:5: error: this 'else' clause does not guard... [-Werror=misleading-indentation]
else
^~~~
dll/3rdparty/libxslt/security.c:365:2: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the 'else'
if (ret <= 0) {
^~
Fixes GCC 8 warning:
dll/win32/rsaenh/rsaenh.c:161:23: error: '%s' directive writing up to 259 bytes into a region of size 235 [-Werror=format-overflow=]
#define RSAENH_REGKEY "Software\\Wine\\Crypto\\RSA\\%s"
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
dll/win32/rsaenh/rsaenh.c:161:23: note: in definition of macro 'RSAENH_REGKEY'
#define RSAENH_REGKEY "Software\\Wine\\Crypto\\RSA\\%s"
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
dll/win32/rsaenh/rsaenh.c:1107:5: note: 'sprintf' output between 26 and 285 bytes into a destination of size 260
sprintf(szRSABase, RSAENH_REGKEY, pKeyContainer->szName);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Fixes GCC 8 warning:
base/setup/usetup/usetup.c:3407:78: error: '%S' directive writing up to 521 bytes into a region of size 200 [-Werror=format-overflow=]
"Setup is currently unable to check a partition formatted in %S.\n"
^~
base/setup/usetup/usetup.c:3406:9: note: 'sprintf' output between 128 and 649 bytes into a destination of size 260
sprintf(Buffer,
^~~~~~~~~~~~~~~
"Setup is currently unable to check a partition formatted in %S.\n"
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
"\n"
~~~~
" \x07 Press ENTER to continue Setup.\n"
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
" \x07 Press F3 to quit Setup.",
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
PartEntry->FileSystem);
~~~~~~~~~~~~~~~~~~~~~~
These are required by latest mingw-w64 stl and we are mixing this with our crt headers.
Define _CRT_NO_POSIX_ERROR_CODES in all modules that define their own constants.
This prevents some errors like this one:
In file included from ../dll/win32/wldap32/ber.c:27:0:
../sdk/include/reactos/wine/debug.h:378:48: error: '__wine_dbch___default' defined but not used [-Werror=unused-const-variable=]
static struct __wine_debug_channel * const __wine_dbch___default = &__wine_dbch_##ch
^
../dll/win32/wldap32/ber.c:29:1: note: in expansion of macro 'WINE_DEFAULT_DEBUG_CHANNEL'
WINE_DEFAULT_DEBUG_CHANNEL(wldap32);
^~~~~~~~~~~~~~~~~~~~~~~~~~
In NtGdiGetGlyphIndicesW function, allocation of zero size had caused fatal failures.
Avoid allocation of zero size in ExAllocatePoolWithTag calls. Optimize for cwc == 0. CORE-12825
- Always include kd64.h
- Change KdpPrompt() prototype to be compatible between KDBG and _WINDK_
- Rename KdComponentTable to KdpComponentTable to prevent a conflict
- Add some functions stubs and global variables
Thanks to the patches author Doug Lyons
and the reviewers hbelusca and Katayama Hirofumi MZ.
The regression was introduced by
0.4.13-dev-24-g
61475b9fed
CORE-12451 CORE-13182 CORE-13196
- CONSOLE: Initialize the additional TrueType fonts cache.
* Fix the font preview when a TrueType font has been selected.
* Refresh the available fonts and the font preview when the selected
code page is changed (Work In Progress), or when the OS pool of
font resources has changed (WM_FONTCHANGE message).
- CONCFG: Implement support for the additional TrueType fonts cache:
the contents of the cache is enumerated under the registry key
HKEY_LOCAL_MACHINE\Software\Microsoft\Windows NT\CurrentVersion\Console\TrueTypeFont
* Add helper functions and macros.
* In CreateConsoleFontEx(), set the mandatory font pitch&family flags
and remove those that we do not support.
* In IsValidConsoleFont2(), update the validity checks and the
documentation links.
- CONSRV: Load/refresh the additional TrueType fonts cache when needed.
This is linked to previous work done on DOS devices creation
in basesrv. If this DWORD is not set (or 0), DOS devices will
be created with an ACL that make the symlink readable by any
and modifiable by any.
With protection mode set, the symlink will be still readable by
any but not modifiable by anyone but the owner.
This should also affect some objects managed by session manager.
By default, on W2K3, that protection mode is set.
- Only include the strictly necessary headers.
- Get rid of the dependency on shell and user DLLs.
- fgetws() gets the string buffer size in number of characters.
- We can use the CRT functions for lengths of the arguments etc.
- The cFileName member of the WIN32_FIND_DATAW structure does not
contain the full PATH to the enumerated file, but only its name.
In order to use _wfopen(), build a full file path out of the
directory part of the file specification and the full file name.
- Simplify a ConPrintf() call to make it "atomic".
- Fix the "confusion" lLineCount vs. lLineNumber vocable in the code.
- Do not emit an extra newline after having displayed the results for
a given file.
- Uppercase the switches for performing the comparisons.
- Send the errors to the StdErr stream.
- Remove trailing whitespace.
This commit contains a complete rewrite of find console utility.
The goal of it was to make the source code easier to maintain
and to add a missing feature (/offline switch).
Additional changes:
* now the program operates on Unicode strings;
* added conutils and shlwapi as program's libraries;
* added IDS_INVALID_SWITCH string into resources;
* modified IDS_USAGE string to include /offline switch
description.
https://ss64.com/nt/find.html was used for reference.
The compiler hits an assert while building a file
The two "officially" supported compilers are working though
This will bring back green status for our CI on GitHub
Some mice give a logical min/logical max which is the maximum move
that they can provide. Those were wrongly detected as absolute mice,
and mouse pointer was mostly stuck at the upper left corner of the
screen.
Fix results from tests, add (last one) gdi batch support for
ExtSelectClipRgn. Left commented out test code in tree this time.
Pass Katayama Hirofumi MZ SelectClipRgn tests. After commit will plug in
the last batch. After 12 years.
See CORE-13817 and CORE-15906.
This will notably bring support for DOS mapping with LUID devices
(not yet supported in the kernel, though).
This also reduces complexity (and thus memory usage) with the "history"
thing. Multiple targets are stored in the link target as MULTI_SZ string.
This fixes regressions introduced with kernel32 fixes/rewrites.
This is also done on Windows for backwards compatibility with Windows 3.x/9x.
But, it's also used (i.e. "required") by some installers, like Doom 3 Demo installer and Battlefield 1942 Single Player Demo installer, for successful opening of their Readme file at the end of their installation!
- Add support for LUIDDeviceMapsEnabled;
- Broadcast proper message in case of device removal;
- Use less memory for strings management;
- Make code a bit cleaner.
by using TEB static unicode string (which is already
preallocated).
Also, properly handle RtlUnicodeStringToAnsiString failures.
Finally, make sure output buffer is properly 0 terminated.
CORE-13182 CORE-13196
- Use the correct character height & width.
- Additions: use StringCch*() when initializing the dialog title.
[CONSRV:CONCFG] Minor fixes.
- When retrieving font characteristics in ConCfgReadUserSettings(),
check for NULL/zero values that indicate that we should use default
ones instead.
- Rename 'dwNumSubKeys' into 'dwNumValues'.
This fixes back journal in ReactOS "at low costs". Indeed,
because write are improperly aligned right now, journaling
just fails.
With that patch, Cc will take care of aligning writes and
journal will be written again. Because flush operations
happen at each and every write to the journal, we expect
changes to land on disk quickly (not as quickly as if
they were directly written). But that's a good trade off
between over engineering and fixing a broken feature.
CORE-15973
This will avoid triggering a FAT repair on
unclean FAT volumes.
If dosfstools.fsck works fine in Linux, its
usage on ReactOS triggers worse corruption
than unclean shutdown.
Given I've no time for debugging this, I
kill it off.
CORE-14638
hid.dll and hidparse.sys must understand the same HID preparsed data,
so use the same code in hid.dll and in hidparse.sys
At the same time, this permis implementation of some HidP_* functions.
Interface between both is not anymore the HidParser_* functions, but
the HidP_* functions and the AllocFunction/FreeFunction/DebugFunctions/
ZeroFunction/CopyFunction.
Add WORLD_XFORM_CHANGED flag to some positioning functions.
Fix IntExtTextOutW function for TA_UPDATECP alignment, so that it can correctly update the current position. CORE-14994
Our On-Screen Keyboard has a manifest and the buttons (the ones with BS_ICON styles) aren't rendered with the specific theme as it should be but instead it takes the classic theme.
The code relies on NM_CUSTOMDRAW notification, which is more intuitive and efficient than doing owner-drawn operations as NM_CUSTOMDRAW allows the controls to use styles whereas you cannot do it on owner-drawn controls.
CORE-15965
This finally fixes our duplicated getopt functions from different sources and gives us an up to date and the most compatible implementation.
isohybrid actually relies on a glibc-specific getopt behavior that we previously hacked into the reactos_support_code.c implementation derived from BSD/mingw-w64.
widl also needs getopt and previously used an even older BSD-derived code.
- The 2nd parameter is already passed in user-mode by GetThreadDesktop().
It is then used by NtUserGetThreadDesktop() as a fallback for console
threads.
- Lookup and validate the thread by using the IntTID2PTI() helper.
- Don't reference the desktop with too many access rights.
- Get rid of the old-school DECLARE_RETURN() & co. macros.
Co-authored-by: Hermès Bélusca-Maïto <hermes.belusca-maito@reactos.org>
It retrieves the handle to the desktop assigned to the specified
thread belonging to a console application, which is going to be
where the input thread of this console is.
Up to CMake 3.3 LINK_DEPENDS was ignored in non-Makefile generators. See also http://www.cmake.org/pipermail/cmake/2010-May/037206.html. Since CMake 3.4 it is no longer ignored, but requires CMP0005 policy to be set to NEW and it requires a full path which was already correctly used on GCC, but not on MSVC builds.
* [CMAKE] Fix for CMP0018
The new behaviour lets us explicitly disabling position independent code, avoiding the need to change undocumented variables
* LINK_DEPENDS is semicolon separated
* [CMAKE] Fix for CMP0005
it seems impossible to have cmake correct escape PROXY_CLSID_IS so move it to the header
unattend.inf only allows to add RunOnce entries, which are in theory only executed on the first boot after installation.
For almost a decade, this worked anyway to start regtest.cmd again after a ReactOS crash, because the registry and thus the deletion of the RunOnce entry wasn't stored on disk in case of a crash.
Thanks to Sylvain for the hint!
As Explorer processes the Run key right after RunOnce and therefore picks up regtest.cmd a second time during the first boot, exit the script just after adding the Run key.
Otherwise, rosautotest.exe would be started twice, with catastrophic results.
This is a rather graceful approach compared to sysreg2's 3 minute timeout before killing and restarting the entire VM.
Since we added autochk for FAT filesystems, the filesystem is often "fixed" after a reset with the consequence that ReactOS doesn't boot up anymore.
The sysreg2 restart code still remains for handling tests causing BSODs.
The callback routine should get the real lengths of the name and description strings in bytes instead of the buffer sizes in characters. This fixes buffer overflows in powercfg.cpl.
- Modify atldef.h for non-ReactOS environments.
- Update ATL testcase project files (from v120_xp to v140_xp).
- #define HAVE_APITEST and use it.
- s/NULL/0/
- Use GetSystemPowerStatus() instead of querying each battery.
- Add an ugly AC power icon. Should be fixed or replaced.
- Display AC power, battery charging, battery depleting and error cases. The AC power case was missing.
- Update some strings because we are no longer using float for the charge percentage.
@Turkish translators: Please check and update the translation of IDS_PWR_CHARGING!
* symtype: set to SymPdb not SymCv (may be for old pdb-files pre VS 6.0)
* flags: return SYMFLAG_PUBLIC_CODE for SymTagPublicSymbol if its a function.
* flags: do not return SYMFLAG_FUNCTION for SymTagFuncion.
Debug symbols (like PDB, ...) could not be found if the direcory of
the module and application differs. Applying this patch they will
be found.
Dbghelp searches the module path too. (See MS docs).
This could happen if BCB was marked dirty previously.
Marking VACB dirty on unpin could lead to a double write of
the VACB, even if clean.
Indeed, now that setting BCB dirty leads to marking VACB
dirty, the VACB can be flushed in between by the lazy-writer.
The BCB state is not reset on VACB flush, contrary to the VACB state.
Thus, on unpin even if the VACB was already flushed, we were
setting back the dirty state, leading the VACB to be flushed again.
This could bring a small performance downgrade. Though it remains
limited since this is mostly used for FS metadata.
Possibly it could lead to metadata corruption, but this is likely
less possible.
CORE-15954
- C4311: pointer truncation from '<pointer>' to '<integer>'
- C4312: conversion from '<integer>' to '<pointer>' of greater size
- C4133: incompatible types - from '<x> *' to '<y> *'
- C4028: formal parameter different from declaration
Eliminate some bugs about font enumeration. CORE-15755
- Add "Microsoft Sans Serif" font substitution.
- Fix and improve the treatment of the nominal font names.
- Split IntGetFontFamilyInfo function from NtGdiGetFontFamilyInfo.
- Add DoFontSystemUnittest function for font system unittest to GDI32.
- Call DoFontSystemUnittest function at CreateFontIndirectW first call.
- Add compatibility with XP and fix some bugs with its rendering engine.
- New Help and Slider bitmaps.
Co-authored-by: Stanislav Motylkov <x86corez@gmail.com>
This fixes the following compiler error:
../dll/win32/advapi32/service/eventlog.c:658:59: error: passing argument 3 of 'ElfrChangeNotify' makes integer from pointer without a cast [-Werror=int-conversion]
dll/win32/advapi32/eventlogrpc_c.h:391:20: note: expected 'ULONG {aka long unsigned int}' but argument is of type 'HANDLE {aka void *}'
- Add and initialize a global power capabilities variable.
- Return the global power capabilities via NtPowerInformation.SystemPowerCapabilities.
- Report the presence of power button, sleep button and lid.
One-character Heisei for CAL_SABBREVERASTRING locale info can be U+337B or U+5E73. And one-character Reiwa U+32FF, U+4EE4, and U+F9A8 are also added. Improve kernel32_apitest testcase. Thanks @ThFabba CORE-15920
_IMPLICIT_LINK_DIRECTORIES
Othrewise, if you link a RC module with a static C library (as done for
fusion DLLs), you get the standard C libraries from GCC. This is not
what we want.
This might have to be done for MSVC builds. Check build.ninja to verify
this.
There is no need to compile our DLLs as shared libraries since we are
managing symbols exports and imports through spec files.
On my system, this reduces the configure-time by a factor of two.
The previous implementation used the resume_handle parameter to return a pointer to the active enumeration context, but resume_handle is a DWORD. To support 64 bit pointers, the enumeration context is inserted into a global linked list and given a unique 32 bit value as identifier for later lookup.
The way the function is implemented, leaking a data structure while the MSDN description does not indicate that, seems a little questionable in general, but that is something that I leave to the original author to investigate.
The Reserved field is ULONG and we need ULONG_PTR. ClassInstallReserved was unused so far. I have no idea whether this is correct, but it cannot be more incorrect than what we had.
This fixes the following compiler errors:
../sdk/lib/crt/stdio/stat64.c:7:13: error: inline function 'release_ioinfo' declared but never defined [-Werror]
inline void release_ioinfo(ioinfo *info);
^~~~~~~~~~~~~~
../sdk/lib/crt/stdio/stat64.c:6:16: error: inline function 'get_ioinfo' declared but never defined [-Werror]
inline ioinfo* get_ioinfo(int fd);
^~~~~~~~~~
../sdk/lib/crt/stdio/file.c:186:5: error: 'init_ioinfo_cs' is static but used in inline function 'get_ioinfo' which is not static [-Werror]
init_ioinfo_cs(ret);
^~~~~~~~~~~~~~
../sdk/lib/crt/stdio/file.c:183:19: error: 'get_ioinfo_nolock' is static but used in inline function 'get_ioinfo' which is not static [-Werror]
ioinfo *ret = get_ioinfo_nolock(fd);
../sdk/include/reactos/rosctrls.h:283:59: error: no matching function for call to 'CToolbar<TItemData>::SendMessageW(int, HWND__*&, int)'
return SendMessageW(TB_SETTOOLTIPS, hWndTooltip, 0);
../sdk/include/psdk/winuser.h:5543:21: note: candidate: LRESULT ATL::CWindow::SendMessageW(UINT, WPARAM, LPARAM) <near match>
../sdk/include/psdk/winuser.h:5543:21: note: conversion of argument 2 would be ill-formed
ReactOS/Windows headers contain lots of structures whose last field is
fieldName[ANYSIZE] or fieldName[1].
Starting with GCC 4.8, gcc considers that accessing items [2], [3], ...
are undefined operation, and may optimize those accesses by removing them!
Add flag -fno-aggressive-loop-optimizations to prevent this unwanted behaviour.
It better captures the intent now in FsRtlIsNameInExpressionPrivate and
fixes a slight overallotion by 4 bytes in FsRtlIsDbcsInExpression.
While at it, use the ANSI_DOS_DOT macro in the Dbcs version.
CORE-15902
CMake >3.2 supports specifying the architecture through the -A parameter,
allowing the logic to be simplified.
Specifying the architecture like this is also required since Visual Studio 2019.
This allows querying volume information without issuing an IRP to the owner device.
The kernel is supposed to already have all the required information to return
to the caller.
Side effect: this allows querying volume information for devices not implementing
IRP_MJ_QUERY_VOLUME_INFORMATION such as null.sys
This fixes opening null device in Python.
Fix based on debugging by Maxim Smirnov in PR #1442
CORE-14551
On-Screen Keyboard has a manifest which is used for visual styles. However, the common controls aren't initialized and as a general rule the controls have to be loaded in the application's handle instance.
On-Screen Keyboard provides a mechanism to launch the application only once, to avoid multiple entry point instances. Such mechanism is based upon mutex objects, although it could happen that the program may end up creating two or more mutexes (a race condition).
CORE-15877
Import Wine commit:
b1b8fb77be
"user32: Add support for navigating a group of radio buttons using a keyboard.
The patch approximates the behaviour observed in the message tests
but still doesn't make the message tests pass without failures.
"
by Dmitry Timoshkov.
See bug report https://bugs.winehq.org/show_bug.cgi?id=16845
CORE-8526
Import Wine commit:
96d0af52eb
"user32: Move the auto radio button group logic from BM_SETCHECK to WM_LBUTTONUP handler.
This patch also changes the logic to get the control style with WM_GETDLGCODE
instead of GetWindowLong to make the message test pass.
"
by Dmitry Timoshkov.
See bug report https://bugs.winehq.org/show_bug.cgi?id=42010
- (ReactOS-only) Fix also the corresponding logic in COMCTL32.
- Enable "New Start Menu" control for all languages.
- Implement the "NoSimpleStartMenu" policy in shpolicy.c and its usage in explorer.
- Hide controls instead of disabling (as in Win XP/Vista).
- Assume that GetTextExtentExPointWPri() uses the same kind of
arguments as GetTextExtentExPointW().
- Add GetTextExtentExPointWPri() declaration to undocgdi.h.
Implement the coordination dialog data saver. This allows OSK to launch using the previous placement values. Such behaviour can be seen with the XP's part of On-Screen Keyboard.
Since ReactOS doesn't actually support non-local WinSTA/WTS capabilities, this seems to be a good way to reduce spam in the log while providing correct behaviour.
Also call Unicode version from ANSI one to reduce code duplication.
- Fixed bug 12.37 13/06/2009 into multiprecision display (unicode only).
- Fixed bug when calculator was closed with statistics dialog open.
- Fixed bug in XrY operator.
- Fixed error into conversion unit.
- Do not append decimal separator if "ERROR" messages are printed.
- Remove call to _tcslen() and use return value of GetDlgItemText() where possible.
Configuration is read from WIN.INI under W32S/Win9x/Me and from the registry
for all NT platforms.
Although with INI file mapping the keys are still saved in the registry
even on NT, I added handling of different platforms for writing directly
the configuration in the right place.
- Fix errors if a theme api is missing.
- Add callback to functions for drawing themed transparent background.
- Fix drawing glitch when theming is applied.
- Redraw on theme change: automatically redraw the window if the
theme is changed while the application is active.
- Colours are now declared though RGB() macro.
- Removed safe DS_SHELLFONT declaration.
- Added support for average of squares and mean of squares into statistical functions.
- pop() and push() functions in RPN engines now work with nodes instead of stack units.
- Moved the POW and SQR operations near the operators.
CORE-12766
- Fix number of digits for IEEE-754 constants.
- Show all available digits in exp notation.
CORE-14871
- Update help correspondingly.
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.
- Add also some validation ASSERTs and simplify the code here and there.
- The installation partition is called "InstallPartition", while the
global "CurrentPartition" is the disk region currently selected in
the partition UI list, on which prtitioning operations are effectued.
- Extend CheckActiveSystemPartition() to use an optional alternative
disk or partition in case the actual system partition (present in the
first disk) cannot be used, e.g. because we don't support writes on it.
- Find or set the active system partition only once, when we start the
formatting stage. If the system partition is to be created in some
non-partitioned space, we create it.
- A file-system check is always scheduled for both the system and the
installation partitions.
- If the system partition was already formatted (which usually means
that it already existed on the system), don't ask for formatting it.
CORE-13205
- For each partition to be formatted, the file-system menu list is
re-created, in order to display only the FSes that are compatible
with the selected partition, and allow to give the user the choice
to keep the existing file-system only if the selected partition was
already formatted.
- Check whether the user attempts to delete the partition on which
the installation source is present, only if it is not "new" and is
formatted.
- Check first whether the current partition is the system partition,
before displaying the CONFIRM_DELETE_SYSTEM_PARTITION_PAGE.
- Make some partitioning functions not dependent on the selected "CurrentPartition".
- Add some sanity checks.
- Improve some of the "is-partitioned" checks.
Similarly to FMIFS this structure should be private. Instead file-system
names are passed to the helper functions, allowing to use the names
returned by the FS drivers. The names are then internally mapped to the
corresponding FS providers.
In particular this allows to handle the "RAW" file-system and to assign
the 'Unformatted' flag to partitions having this FS.
Finally this helps us refining the checks performed to see whether the
current "active" system partition uses a supported file-system.
Add fix ups to PolyPatBlt and freetype.
TextOut supports 580 characters w/o Dx and Dx at a max of 193 characters
with Dx*1, both if offset is 0.
Noticed a lot of over drawing with start menu from ComCtl32 SB_SETTEXT
going through User32 DrawText/Worker ExtTextOutW. Explain why it might
be slowing down.
If issues arise, commenting out lines in win32ss/include/gdi32p.h can
help.
Example:
else if (Cmd == GdiBCTextOut) cjSize = 0;//sizeof(GDIBSTEXTOUT); <----
this one most of all.
else if (Cmd == GdiBCExtTextOut) cjSize = 0;//sizeof(GDIBSEXTTEXTOUT);
Test results:
https://reactos.org/testman/compare.php?ids=66260,66262
Separate TextOut for internal use.
Use flTextAlign internally.
Not sure why GetDeviceCap was used for determining rendering. SVN commit
logs are unavailable. Include code from wine that might make more sense.
MSVC was previously given a "result" variable to copy the fscale result from st(0). This led to another "fld" FPU stack push at the very end without popping the source value from the FPU stack.
Moreover, this copy isn't even needed: A simple "fstp st(1)" at the end pops an element from the FPU stack while effectively storing the result in st(0), the register used for returning a double value.
This problem didn't affect GCC, as it is only given the "fscale" instruction and does all necessary stack operations itself.
However, looking into the CRT sources, I found many other i386 implementations with inline assembly suffering from the same problem.
Fortunately, they have been replaced by pure assembly implementations a while ago, so it's time to finally remove them.
ldexp would have also been a candidate for a pure assembly implementation, but the required check for NaN and setting errno (verified on Win2003) already outweighs the benefits.
And we cannot just do a NaN check with FUCOMI as this is an i686/pentiumpro instruction while we're still targeting i586/pentium.
I'm also using this opportunity to clean up the ldexp.c header and only put in the remaining contributors as returned by "git blame".
Thanks to NightWolve1975 for reporting the problem! (https://twitter.com/nightwolve1975/status/1099042477531643912)
This stub is responsible for providing the UUID seed to the kernel.
For now, it generates a purely random seed - MAC address support
is to be added.
Because it's purely random seed and local the machine (not tied
to the MAC address) it's made so that ExUuidCreate keeps returning
RPC_NT_UUID_LOCAL_ONLY. It won't fix failing test for now.
Nota: this service shouldn't exist as it. It should be implemented
though rpcss DLL for both rpcss service (network service) and
dcomlaunch service (local system service).
Because rpcss is EXE based and wine-synced for now, I prefered
stubbing a new service. This will have to be changed at some point.
IDD_BROWSE_FOR_FOLDER displays text controlled by the calling applications in
IDC_BROWSE_FOR_FOLDER_TITLE. Its old height could lead to text truncation
for existing 3rd party apps.
The new height value for IDC_BROWSE_FOR_FOLDER_TITLE
was inspired by IDC with the same from dialog IDD_BROWSE_FOR_FOLDER_NEW.
I could change all languages to the same height,
as dimensions of the dialog and surrounding controls were same.
- Implement the save/load handler for this feature
- Check the "Use Click Sound" item state accordingly depending if it's enabled or disabled
- Use PlaySoundW() to play the wave sound file from resource
- Add the WAV sound click file. The work is made thanks to Midori Mizuno
We previously only gave the device a hard-coded amount of time to respond,
which could lead to interpreting the contents of uninitialized memory as
a response. This would lead to an unreasonably large number of audio function
groups being detected.
A KSEMAPHORE mirrors what Haiku uses here, though it may not be the optimal
synchronization primitive for this case under Windows.
This protects against crashing in case of faulty/malicious hardware,
but also works around a bug in HDA_SendVerbs that causes it to return
invalid data, thereby suggesting more groups than are actually present.
This will avoid 0-sized allocations, or -1-sized allocations.
So far, it's maxed by hard value stored in TCPIP.sys. I believe
this is not right and would deserve a true fix
* Add all wallpapers to the ISO and can look how enable the module in ReadMe.txt
* Remove blanks in name.
* Moving to folder with same wallpaper style.
* Adding and updating CMakeLists.txt.
* Update ReadMe.txt.
* 1 "warning: variable 'Status' is used uninitialized whenever 'if' condition is false [-Wsometimes-uninitialized]"
* 2 "warning: comparison of unsigned expression < 0 is always false [-Wtautological-unsigned-zero-compare]"
* Use a consistent type for "i" and fix a comment
* Also update licence header.
CORE-14306
Also stub the related functions the IOCTL will return, namely: WSAID_ACCEPTEX,
WSAID_CONNECTEX, WSAID_DISCONNECTEX, WSAID_GETACCEPTEXSOCKADDRS
This will allow Steam client not to fail in an ASSERT
related to the Steam client PlatformSocketsInit function.
Even though, this doesn't fix Steam running on ReactOS since its UI requires
a Windows 7 API from Kernel32: SetThreadErrorMode().
Nota: its crash got fixed thanks to the Wine sync done recently.
CORE-14966
Prevent nslookup.exe from crashing when executed in a ROS VM with no network
interfaces. This is due to a NULL pointer dereference occurring if
`GetNetworkParams` in `main` fails with an error other than
`ERROR_BUFFER_OVERFLOW`. In this case, `pNetInfo` remains initialized to
NULL, causing `strncpy` to crash.
Clang-CL does not support #pragma function like CL. The alternative to this approach is to disable all intrinsics for this single source file, but that is not trivial to do, as it can only be done by removing a compiler switch, which we have no CMake support for. Therefore this solution is simpler and as good.
- Bugcheck VIDEO_DRIVER_INIT_FAILURE in case initializing video fails.
- Bugcheck WIN32K_INIT_OR_RIT_FAILURE (Windows-compatible) in case
the USER subsystem version does not match.
This is needed in order to avoid an infinite recursive loop between
disk!UpdateRemovableGeometry() and ntos!IoReadPartitionTable().
This does not happen with NT5+ disk_new.sys because it doesn't call
IoReadPartitionTable() in that situation.
- Localize error message and shell file extension description.
- Added Hebrew resources.
- Added Icon.
The icon is the Tango icon named "certificate". If anybody has a idea for more fitting icon please recommend it.
Follow up of #1343
CORE-15736
- Within the function's body code, check the status values returned by the called functions.
- Change the BuildInstallPaths's function type to NTSTATUS instead of VOID (and check the status of InitDestinationPaths() as well.
- Add the respective standard keyboard resource.
- Implement the standard/enhanced keyboard dialog switcher.
- Implement save/load handler.
- Move the warning box conditinal check (bShowWarning) to wWinMain application entry point (this will prevent the modal welcome box from displaying each time the user switches into another dialog).
CORE-14428
MSBuild, via the VS IDE, uses response files when calling CL or LINK.
We cannot specify a custom response file on the linker command-line,
since specifying response files from within response files is forbidden.
We therefore have to pre-process, at configuration time, the linker
script so as to retrieve the custom linker options to be appended
to the linker command-line.
"configure.cmd VSSolution" was failing due to add_asm_files macro
expansion. It was using the same 'freeldr_linker_file' variable name
as a temporary filename base twice. This patch uses '_target'
parameter to make those variable names differ.
"Release" builds should not BSoD on 'Freeing pool - invalid tag specified' case,
this is compatible with Windows behaviour.
Nothing changes otherwise concerning the "Debug" builds, where pool-tag mismatch
triggers a BSoD as expected.
CORE-15446
This reverts commit 0.4.12-dev-347-g
75a80ec823
by explcit request of Giannis Adamopoulos.
I double-checked and confirm, that
this revert fixes both regressions CORE-15595 & CORE-15696.
On the downside the revert will intensify again the regression CORE-13278
(will make the text flash again as well).
Giannis Adamopoulos mentioned earlier, that using double-buffering
eventually could be a valid way out of CORE-13278 instead.
- Correctly probe the FileName user pointer before capturing its contents.
- A paging file name with zero length is invalid too.
- Remember (TODO!) in the code that the lower bounds of the Safe*Size values
need to be checked too!
Running out of pool is likely at least during kmtest:ExPools.
There is a chance of crashing when dereferencing these null pointers -- but
worse, there's also a chance of overwriting the IVT or BDA if a VDM BIOS
call is in progress, which can lead to crashes in non-obvious places later.
- Multiboot support added back in FreeLdr MSVC builds so that it can be
booted using e.g. GRUB. See CORE-15563.
- Re-introduce the disk drive mapper code.
As the MSVC linker alone doesn't permit such control, the file uses
ASM language (C can alternatively be used), together with extra linker
command-line switches.
It is pre-processed 3 times: first for generating the ASM code, second
for the C code and the third time for generating the linker response
file.
In our case, the ASM code defines the __bss_start__ and __bss_end__
symbols that allow us to find the limits of the .bss section (which is
by the way automatically appended to the .data section by the MSVC linker).
The C code is used to specify the list of linker switches that can be
passed through the `#pragma comment(linker, ...)' directive (the can be
alternatively specified in the linker response section).
Finally the linker response section contains all the linker switches
that cannot be specified with the `#pragma comment(linker, ...)'
directive.
Using all this we can recycle the BSS initialization code, that has been
written originally for GCC only, also for the MSVC builds.
Also, remove the outdated .text16 section merging.
- Whitespace;
- Explicitly use string pooling (option "/GF"), so that even if it's
used globally in the main ReactOS CMAKE file or not, we still have it
there, because we are quite size-constrained;
- Rename the source lists (OK, we understand it's _COMMON_ ...);
- Remove redundant double inclusion of linuxboot.c;
- In the 'freeldr_common' static library, compile the ASM code first.
Modify the .code16 macro so that the 16-bit code segment can get a
chance to be correctly merged with other (possibly 32-bit) code
segments, without keeping generating multiple .text segments with
different attributes and generating the corresponding LNK4078 warning.
This fixes the warning when compiling NTOSKRNL on MSVC:
"v86.S.obj : warning LNK4078: multiple '.text' sections found with different attributes (C0520040)".
- Open a separate mixer instance for the page, so that MM_MIXM_CONTROL_CHANGE notifications can be received.
- Keep line and channel configurations in a global struct.
- Handle MM_MIXM_CONTROL_CHANGE and WM_HSCROLL notifications.
- Simplify failure path for DeviceName query failure
- Don't make stable query failure fail the whole function call
Based on a patch by Vadim Galiant
CORE-15550
With this commit, we now use a forked version of MESA which only supports OpenGL 1.1, like the windows implementation does.
It exposes :
- The same pixel formats
- The same set of extensions
- Nothing more
All of this without taking 10% of your build time.
If you need a more modern option, look at the MESA package from Rapps, which is (and must be) maintained outside of this code tree.
CORE-7499
* The previous version was overcomplicated and broken and therefore disabled.
* The new version also enforces NX protection on x64.
* Now that protecting works, also protect the boot loaded images.
Implemented the following actions: stick the window on the left/right or maximize it,
with the following shortcuts:
- Win key + Arrows;
- drag to left/right screen border;
- double-click on top/bottom.
CORE-12845
Previously, we would treat the input as unsigned coordinates, this is wrong!
If the coordinates are invalid, we try to find a focused or selected item.
The center of this item is where we'll show the menu.
When there is no item, we default to 0,0
- Increase the height of the trackbar in order to make it fully visible.
- Remove the selection range code.
- Set a range from 0 to 500 and fix the volume calculations accordingly.
- Set a tick frequency of 50 and a page size of 100.
- Play the default sound on TB_ENDTRACK notifications.
- Always use string pooling when building: this helps reducing the
size of the binaries due to string redundancy coming from the usage
of __FILE__ / __RELFILE__ in the debugging helper macros. Note also
that GCC builds use string pooling by default.
- Use suitable add_compile_flags() command.
- Add some explanative comments for some settings.
Some numbers (obtained with my local builds):
Before / After => Reduction
===========================
freeldr.sys : 443 KB ( 453.632 bytes) / 364 KB ( 372.736 bytes) => ~ 18%
win32k.sys : 1877 KB (1.922.048 bytes) / 1562 KB (1.599.488 bytes) => ~ 17%
ntoskrnl.exe : 2253 KB (2.307.072 bytes) / 1902 KB (1.947.136 bytes) => ~ 15.6%
kernel32.dll : 3008 KB (3.080.192 bytes) / 2906 KB (2.975.744 bytes) => ~ 3.4%
Previously, we would just stick the menu on the edge of the screen.
We should actually try to flip the menu around the point of origin,
and only when that fails move it to the edge of the screen.
CORE-15001
CORE-9037
This allows avoiding one of the previous implementation limits:
leaked IRP not queued to a thread are now totally visible since
we look directly in the memory pool.
For now, it allows searching for pool allocations in
both paged and non paged pool.
It is based on Andreas Schuster work to identify POOL_HEADER
structures.
- Add a check in co_UserCreateWindowEx() for parentless windows,
that checks the default layout direction; if it's LAYOUT_RTL
add the WS_EX_LAYOUTRTL flag to the extended window styles.
- Make the internal routine accepting also LAYOUT_LTR as a value for SetProcessDefaultLayout().
Limit receiving value to LAYOUT_ORIENTATIONMASK (and not just LAYOUT_RTL)
or LAYOUT_LTR, as per written in:
https://docs.microsoft.com/en-us/windows/desktop/api/winuser/nf-winuser-setprocessdefaultlayout
Now all the applications that call SetProcessDefaultLayout() to mirror the layout get mirrored.
This is based on Wine.
Fix EnumFonts, EnumFontFamilies and EnumFontFamiliesEx functions.
The charsets of the font substitutes are also enumerated.
Delete meaningless codes. CORE-15558
- Introduce the AddReactOSArcDiskInfo() helper to get rid of the
duplicated reactos_arc_disk_info usage scattered amongst different
files.
- GetHarddiskInformation(): If we cannot read the disk that has been
enumerated by the BIOS, return a default identifier string.
- GetHarddiskInformation(): The last character of the disk identifier
string corresponds to whether its MBR is "valid"/has the 0xAA55
signature (use 'A') or whether it's not (use 'X'). Tested on Windows.
- Split PcInitializeBootDevices() in two: the first part of this
function that enumerates hard disks from the BIOS is now moved into
EnumerateHarddisks().
- The 'ValidPartitionTable' field of the ARC disk signature records must
be initialized with correct value that is specified in the different
AddReactOSArcDiskInfo() calls.
This change is similar to what is done in UserChangeDisplaySettings()
after changing screen video mode.
This allows e.g. having windows to be correctly maximized during
2nd-stage GUI setup. To compare and reproduce during 2nd-stage GUI
setup, open cmd.exe (Shift-F10) and from there a GUI app, e.g.
regedit.exe, and maximize it. Observe the limits used by the window.
[CMDUTILS/AT] Fix missing translation strings in certain files.
[NOTEPAD] Fix SUBLANG code to brazillian.
[RAPPS] Fix missing translation strings in certain files.
[FDEBUG] Fix translation string ID.
[CPL/INPUT] Fix missing translation strings in certain files.
[ACPPAGE] Fix incorrect resource IDs.
[NETSHELL] Fix incorrect resource IDs.
[DEVMGR] Fix missing translation strings in certain files.
[LSASRV] Fix missing translation strings in certain files.
[RASDLG] Fix missing translation strings in certain files.
[SHELL32] Fix missing translation strings and incorrect resource IDs.
[TAPIUI] Fix missing translation strings in certain files.
[WINFILE] Fix incorrect resource IDs.
[NTVDM] Fix missing translation strings in certain files.
[USERSRV] Fix missing translation strings in certain files.
[BROWSEUI] One more missing string.
[FLTMC] Fix missing translation strings in certain files.
Detected using the TransDiffer tool (early alpha).
This doesn't include everything anymore, but I wanted to get the PR out of the way.
* [WIN32K] Fix handle calculation in DbgGdiHTIntegrityCheck
* [NOTEPAD] Fix MSVC warnings
* [PSDK] Simplify *PROC definitions in windef.h
* [VIDEOPRT] Don't try to use NtVdmControl on x64
* [FREELDR] Fix some macros
* [CRT] Make qsort 64 bit compatible
* [NTOS] Use #ifndef _WIN64 instead of #ifdef _M_IX86 around C_ASSERTs
* [FAST486] Fix 64 bit warnings and change DWORD to ULONG, so it can be used in kernel mode
* [APPHELP_APITEST] Fix 64 bit issue
This shouldn't be used only for console applications but can potentially be used by any application to receive shutdown notifications.
MSDN provides more information in the documentation for SetConsoleCtrlHandler. Our services.exe also expect to receive shutdown notifications in this way.
UserClientShutdown will never be called for csrss so we don't need to have a check for that. The existing check was broken and wasn't doing anything anyway.
Handle processing winlogon by doing nothing so that consrv won't be bothered about it.
This helps when e.g. changing the resolution on the Dell Latitude D531,
which reports that it supports large resolutions (e.g. 1920x1440x32 and
others larger than 1024x768x32) but fails to apply these.
This usually happens because PDEVOBJ_pSurface(), and more precisely
ppdev->pldev->pfn.EnableSurface(), fails for these resolutions.
- PDEVOBJ_bSwitchMode(): Set the new video mode, or restore the original
one in case of failure + release the allocated ppdevTmp if previous
calls fail. Also unlock in reverse order of locking order.
- UserChangeDisplaySettings(): In case PDEVOBJ_pSurface() fails (but has
reverted the original video mode), we still need to refresh the
display since the display may have been messed up.
- NtUserOpenInputDesktop: Don't crash if there is no input desktop yet
- NtUserOpenInputDesktop: Fail if the process doesn't belong to the interactive window station
- NtUserCreateWindowStation: Clear error on success
- DesktopWindowProc: Use UserOpenInputDesktop to get a handle to the input desktop
This is far from perfect, and totally doesn't match the
WinDBG way of doing it. Instead of browsing pool to find
matching 'IRP' tags, we just browse all the processes
to find the queued IRP. This requires the IRPs to be queued,
obviously, and will make us miss the leaked IRPs, for instance.
Proper way to do it would be to implement !poolfind and then
rely on its implementation to find our IRPs.
Perhaps later ;-)
- Add missing parameter to a DPRINT.
- Print pointers using %p instead of %x.
- Move all executable code behind a functions initial DPRINT.
- Do not use non-trivial initializers.
- Check for NULL instead of using !.
- Add missing parentheses.
Also, show that querying owner information about a connection owned by a service
returns the service name, and not the image nor its path.
Obviously, that fails in ReactOS.
Partially implement font/text rotation by LOGFONT.lfEscapement and improve the rendering image. If the angle was not a multiple of 90 degrees, then the background won't be filled. CORE-11848, CORE-15319
This fixes services not being found.
With that commit and the previous one, I_QueryTagInformation is now fully
functionnal and properly returns service name for a specific tag
This is wrong and messes up with internal structures (such as TEB).
For the two needed functions which are "NT6+", they were exported
with NT6, but already existed in W2K3 SP2.
This might fix random bugs
CORE-15526
the type of device it should return must not be FILE_DEVICE_CD_ROM, but must be the actual underlying one!
This fixes recognizing *removable* USB keys as disks and not CD-ROMs.
CORE-13184
This may look strange, since this buffer is originally allocated using
the TAG_IO_NAME tag. However, it happens that file-system drivers are
allowed to re-allocate this buffer: this is what the MS' open-sourced
CDFS driver does, see e.g. CdCommonCreate() and CdNormalizeFileNames()
in cdfs/create.c .
This fixes a pool tag mismatch 'mNoI' != 'nFdC' BSOD when resources
are freed when closing a file that has been opened with a relative name
on a CDFS-mounted volume.
* Add an NDK header to define INIT_FUNCTION/INIT_SECTION globally
* Use _declspec(allocate(x)) and _declspec(code_seg(x)) on MSVC versions that support it
* Use INIT_FUNCTION on functions only and INIT_SECTION on data only (required by MSVC)
* Place INIT_FUNCTION before the return type (required by MSVC)
* Make sure declarations and implementations share the same modifiers (required by MSVC)
* Add a global linker option to suppress warnings about defined but unused INIT section
* Merge INIT section into .text in freeldr
By default, we still fallback to FAT if nothing asked, or if there
is an invalid input.
0 is FAT, 1 is BtrFS. This can be grown as soon as we add more IFS.
[REG][REACTOS][EXPLORER][APPWIZ][CONSOLE][INETCPL][INPUT]
[INTL][OPENGLCFG][TIMEDATE][DEVCPUX][COMCTL32][COMDLG32]
[MSGINA][NETCFGX][NETID][SETUPAPI][SHELL32][SYSSETUP]
Update Simplified and Traditional Chinese translations.
For now, this is just a split between scan and flush that
were both done during lazy scan previously.
Lazy scan shouldn't perform any write operation, but only
queue a write behind operation.
Our implementation is far from the original, as it seems
our lazy scan should queue a write behind operation per
shared cache map. Right now, we only perform global
operation.
@Translators: Please start translating messages from netmsgmsg.mc and errorcodes.mc (located in sdk\include\reactos\mc). They will be used instead of local strings.
This will avoid corruption when a file size is little grown and read afterwards.
Up to now, FSD where reading 0es instead of expected data, causing corruption.
This fixes MS FastFAT not being able to mount a FAT volume in ReactOS, corrupting
the FAT.
This also fixes the CcSetFileSizes kmtest tests.
This is based on a patch by Thomas Faber.
CORE-11819
* [INF] Update Traditional Chinese translation.
* [ROSAPPS] Update Chinese translation.
* [FONTSUB] Add Traditional Chinese translation.
* [FONTSUB] Add Traditional Chinese translation.
* [FONTSUB] Update Simplified Chinese translation.
* [ROSAPPS] Update Chinese translation.
* [FONTSUB] Update Simplified Chinese translation.
Import Wine commits by Nikolay Sivov:
* d1e618d31da usp10: Use glyph index consistently in ScriptGetGlyphABCWidth().
* 8d018d8d1da usp10: Make sure metrics cache is used with glyph indices only.
* eda3a3e37da usp10: Remove extra parentheses.
Fixes font display in UltraISO and GOG installers.
See also https://bugs.winehq.org/show_bug.cgi?id=44410
This incorrect behavior was leading to a call at too high IRQL for paged code.
This was triggered by MS FastFAT.
ReleaseFromLazyWrite call was already correctly called to that regard.
CORE-11819
Adds the 'Red, White, and Blue' Color Scheme used by the Classic Theme and adds localization for all the color schemes into themeui.dll.
Translators, please check the language files for potential copypasta errors.
* [BROWSEUI] Update Traditional Chinese translation.
* [BROWSEUI] Update Simplified Chinese translation.
* [SHELL32] Update Traditional Chinese translation.
* [REGEDIT] Update Traditional Chinese translation.
* [BROWSEUI] Update Traditional Chinese translation.
- The converter is code7bit I made. code7bit: https://github.com/katahiromz/code7bit
- This converter converts non-clean characters in string literals.
- However I couldn't erase a ton of codepage C4819 warnings.
- CORE-15447
You can edit an European source file as follows:
- Right click the file in the Solution Explorer.
- Select "Open With...".
- Choose "C++ Source Code Editor (with encoding)".
- Choose "Central European (Windows) - Codepage 1260".
Also, probe the service name when unloading a driver if called from
user-mode. This will avoid that userland applications can trigger an
invalid read in the kernel (and thus, a BSOD).
CORE-15468
- Require the user to have TCB privilege for using this function.
- Probe and capture the user-provided string (and avoid usermode-triggered BSODS ;-)
- Allocate the OEM-converted string in *NonPagedPool* because we are
going to transmit the buffer to BOOTVID.
- Add UserCreateSystemThread function that will signal csrss to create a new system thread.
- NtUserCreateWindowStation: Create the raw input thread and the desktop thread when the IO window station gets created.
- IntMakeHungWindowGhosted: Create the ghost system thread that will own all ghost windows.
- Let the raw input thread manage the window station of csrss.
[USERSRV] Remove system threads creating hack
- Implement SrvCreateSystemThreads
- Don't create the system threads in UserServerDllInitialization.
- NtUserSetInformationThread: Stub UserThreadUseActiveDesktop and UserThreadRestoreDesktop
- Properly mark the first thread that enters win32k belonging to csrss. At this point we assume that since gpepCSRSS isn't initialized yet, it probably is the first thread.
[WINSRV] Use NtUserSetInformationThread to set the current desktop when needed
-When csrss needs to use user32 or enter win32k, it first needs to assign the current thread to a desktop.
* [SYSSETUP] Update Simplified Chinese translation.
* [SYSSETUP] Update Traditional Chinese translation.
* [SHELL32] Update Traditional Chinese translation.
* [VCDCONTROLTOOL] Update Simplified Chinese translation.
* [VCDCONTROLTOOL] Update Traditional Chinese translation.
Fixes MSVC warnings:
sdk\include\crt\xmmintrin.h(33): warning C4162: '_mm_getcsr': no function with C linkage found
sdk\include\crt\xmmintrin.h(35): warning C4162: '_mm_setcsr': no function with C linkage found
addendum to 0.4.10-dev-502-g
d559ca9c98
That copy-paste was done once intentionally to keep our imports similar to 2003.
It's worth reading the conversation in PR #748: Getequ seems to be ok with this.
The ideal solution most likely would be to instead *statically*
link against RegLoadMUIStringW() from advapi32_vista when WINNT=0x502.
For now the stuff runs again at least.
I left the func existing in advapi32_vista.dll.
- Implement SamConnect5() and let the other connect functions call SamrConnect5().
- Add missing and improve incomplete TRACE messages for all Samr functions.
Using locked operations (insertion & removal) on the list of queued
devices installations is necessary, because these operations are done
concurrently by two different threads: PnpEventThread() and
DeviceInstallThread().
Addendum to commit b2aeafca (r24365).
Do not enable it yet, as it doesn't work in ROS for the moment :-(.
Its place in tree is not optimal (it should be with disk/class/etc.),
but I prefer keeping it close to actual driver for now.
All the work has been done so that it compiles and links with ReactOS
SDK though.
Some initial work is done, all testcases pass now.
However a lot of the patch format is still unknown / not working,
so do not expect anything other than the apitest to work yet.
The next thing to do now, is to gather input, output and patch files,
and analyze how the compression and patching works on bigger /
more complex files.
CORE-15417
- Add checks for GPT disks and either fail or ignore the disk,
depending on the operation being executed.
[USETUP][REACTOS] Display the disk style more accurately.
* Speed up the script by using the fallback by default
* Suggest what cmake file to edit
* Add support for x64 modules (addresses not updated yet)
CORE-14923
This test will fail on everything < W2K3 SP2. In spite of what a wide spread
rumor says, MS seems to have added that feature quite lately, but not starting
Vista! ;-)
This usually happens when the build dir is a subdir of the source dir. It is now detected during runtime, using the length of the relative path from the build dir to the source dir as the length of the prefix to skip, if __FILE__ starts with a '.'.
Also fix the escaping of REACTOS_*_DIR defines. It was gracefully fixed up by CMake, but resulted in broken syntax highliting.
CORE-14839 #resolve
This is just an open sourced version of the tool provided by MS along
with the driver.
It will avoid users fighting with the CLI tool (that was doing the job
though ;-)).
Looks like that: https://twitter.com/HeisSpiter/status/1072482763348107264
Nota 1: it doesn't have an icon
Nota 2: code may be ugly, I'm not a umode guy! ;-)
CORE-7749
- Correctly insert discovered partitions in sorted order of StartSector,
and verify that they do not overlap (-> check for broken partitioning).
May help for CORE-10898.
- Use the correct reported partition numbers that may be modified after
partitioning changes, and that need to be used when opening
\Device\Harddisk'M'\Partition'N' files. This is achieving by
retrieving the returned value of the IOCTL_DISK_SET_DRIVE_LAYOUT call.
Distinguish them from the "on-disk" partition numbers that are the ones
that enumerate the partition in partition-table order (and is the order
known by e.g. the BIOS), and that should be used to construct the
destination ARC path.
May help for CORE-4870, CORE-13205.
- Simplify a lot of duplicated code by using helper functions.
This avoids performing a double-free (even though that's hidden by the
fact we use lookaside allocations for VACB), and it avoids freeing
a memory address at an uninitialized address.
We don't care about references here, the VACB was just allocated, never
linked and we're its only user.
CORE-15413
Update the name of the Romanian Keyboard layout in localized
resources by renaming the existing one as legacy and adding
the new one as standard.
Use the same resource string ID as the one used in input.cpl from
later versions of Windows that do have kbdrost.
This improves the Windows binary compatibility of kbrost's setup.
CORE-15056
Define the Romanian Standard Keyboard layout, as defined by
Romanian Standards Association (ASRO). This also renders the
existing Romanian Keyboard layout as Legacy.
CORE-15056
It contains all the Unicode characters that can be formed with a
given dead key (i.e. combining character). Keyboard layouts with
dead keys may find these useful.
CORE-15056
Not much has changed, several options were added, but left at their default values.
However, if search problems persist after Doxygen 1.8.13 generated the documentation for the first time, we can play around with EXTERNAL_SEARCH.
In the struct passed to SHCreateFromDesktop the fourth field is the parameter that later on will be passed to ShowWindow. Initialize it properly so that when we run filrbrowser.exe in win2k3 the main window won't be hidden.
-Rename StartMenuBtnCtxMenuCreator to CStartMenuBtnCtxMenu_CreateInstance.
-Rename CreateStartMenuSite to CStartMenuSite_CreateInstance.
-Remove a couple of definitions that were not used.
NtTerminateProcess just queues an apc in every thread of the target process which in turn kills each thread. We need to wait so that all processes have enough time to exit before shutting down.
Extracted from the pci.ids database at https://pci-ids.ucw.cz/ from 2018-11-21 03:15:02 .
Maintained by Albert Pool, Martin Mares, Michal Vaner, and other volunteers from the PCI ID Project.
Before taking any default action if there is no timeout, check whether
the supplied key filter callback function may handle a specific user
keypress. If it does, the menu timeout is cancelled.
This allows e.g. handling F8 press for displaying boot options even when
the timeout is zero.
CORE-14046
Matching the existence of Eclipse CDT ignore rules, .vscode should also be added to .gitignore
Visual Studio Code is always eager to create settings.json file there.
- Cache the RootFcb so that its cleanup can be handled separately
during dismounting.
- Force volume dismount at cleanup if the VCB_DISMOUNT_PENDING flag
is set.
- Actually dismount a volume if its VCB has been flagged as not good,
or if we force dismounting.
NOTE: In their *CheckForDismount() function, our 3rd-party FS drivers
as well as MS' fastfat, perform a comparison check of the current VCB's
VPB ReferenceCount with some sort of "dangling"/"residual" open count.
It seems to be related to the fact that the volume root directory as
well as auxiliary data stream(s) are still opened, and only these are
allowed to be opened at that moment. After analysis it appears that for
the ReactOS' fastfat, this number is equal to "3".
- On dismounting, cleanup and destroy the RootFcb, VolumeFcb and the
FATFileObject. Then cleanup the SpareVPB or the IoVPB members, and
finish by removing the dismounted volume from the VolumeListEntry
and cleaning up the notify synchronization object and the resources.
- During dismounting, and on shutdown, flush the volume before
resetting its dirty bit.
- On shutdown, after volume flushing, try to unmount it without forcing.
- Release the VCB resources only when we actually dismount the volume
in VfatCheckForDismount().
- Initialize first the notify list and the synchronization object,
before sending the FSRTL_VOLUME_MOUNT notification.
- If we failed at mounting a volume but its VCB's FATFileObject was
already initialized, first call CcUninitializeCacheMap() on it
before dereferencing it.
- Send FSRTL_VOLUME_LOCK, FSRTL_VOLUME_LOCK_FAILED and
FSRTL_VOLUME_UNLOCK notifications during volume locking (and failure)
and volume unlocking.
- Flush the volume before locking it, and clean its dirty bit if needed.
NOTE: In addition to checking for VCB_CLEAR_DIRTY, we also check for the
presence of the VCB_IS_DIRTY flag before cleaning up the dirty bit: this
allows us to not re-clean the bit if it has been previously cleaned.
This is needed for instance in this scenario:
- The volume is locked (it gets flushed and the dirty bit is possibly cleared);
- The volume then gets formatted with a completely different FS, that
possibly clears up the first sector (e.g. BTRFS ignores 1st sector);
- The volume is then dismounted: if we didn't check whether VCB_IS_DIRTY
was set prior to resetting it, we could attempt clearing it again! But
now that the volume's filesystem has been completely changed, we would
then try to modify the dirty bit on an erroneous position on disk!
That's why it should not be touched in this case during dismounting.
- The volume is unlocked (same comment as above), and later can be
detected as being BTRFS.
addendum to 0.4.11-dev-673-g
04e1263921
Actually this control must not statically consume width of 170.
There are two different versions of that dialog.
One has a "change"-button. In this case the edit must be only 100 wide.
In case we have the version without the change button, we resize it to have
170 width (same as its neighbor).
This commit is shared work of
Stanislav Motylkov, Giannis Adamopoulos and Joachim Henze.
Commit supersedes PR#1056.
This fixes the API not to return the current TCP connections while
enumerating them (and thus, netstat not displaying them).
It seems we return garbage for now though
Always perform the partition validity checks when creating new
primary/extended/logical partitions, and also when a new partition
is automatically created when unpartitioned space is selected for
ReactOS installation.
CORE-12246
We first check if a socket is listening before
checking whether it has connections. This allows
properly returning listening address.
Furthermore, if it's listening, properly return
status so that it displays nice in netstat.
Result: https://twitter.com/HeisSpiter/status/1064245622323200000 :-)
Use a central compile option like we do in the rest of the shell so that when testing in windows we can switch using the one in the current dll or the one registered in the system.
Note that not only CLSID_ConnectionManager isn't implemented in netshell, it can't even be instantiated as CLSCTX_INPROC_SERVER because it is implemented by the netman service
That hack (cf. previous commit) was used to restore all
the saved connections. But, because it was implemented
in the wrong place, each time the MPR DLL was attached,
it was trying to restore connections.
With that commit and the previous one, now, ReactOS has
a correct behavior (it tries to bring back connections
only at session opening) and will reduce "workload" for
applications attaching MPR dll.
CORE-15310
In Windows XP, both in On-Screen Keyboard and in Magnify there is a welcome box which is shown upon startup of the application. This feature is already implemented for Magnify in ReactOS which however in On-Screen Keyboard is still unimplemented.
Purpose:
When doing a PXE BootCD installation, you may be left without networking, usb and a cd drive.
This adds an easy way to include a few or more files to the installation which are then present on the disk after the installation and which does not require any change to ReactOS source code/build script files (and no manual editing of boot/bootdata/packages/reactos.dff.in).
Proposed changes:
Adds a new modules/3rdparty folder which lets the files end up in C:\ReactOS\3rdParty ready for use.
* [MSPAINT] Fix scrollbar presence on canvas/view fitting
This is a small fix for a scrollbar related edge case functionality exhibited in mspaint (and in a lesser degree also elsewhere in instances of) scrollbar controlled content. As known, the scrollbars presence is given by the size ratio between the accommodating outer frame and the scroll-controlled contents inside it, and it works for the most part. There is, however, a case where the presence of both toolbars is kept even when the contents would fit the frame (if not for the scrollbars themselves that increase the size of the contents' area). Please take the current commit as a demo fix and a provisional correction for mspaint (as I think that the proper fix should reside in common controls).
* [MSPAINT] Fix a small mistake and improve readability in UpdateScrollbox()
Although with no visible impact on testing, I've used a wrong constant related to vertical scrollbox width in one instance (SM_CXHSCROLL instead of SM_CXVSCROLL), and this fixes it. Also, prefer using constants, to improve a bit readability and performance (by avoiding repeated function calls).
- Simplify the volume-deletion code in RawCheckForDismount().
- Fixes the OpenCount check in RawClose(): the VCB mutex must be
released when the volume has not been dismounted, either because
OpenCount != 0 or because RawCheckForDismount() returned FALSE.
- Explicitly use VCB_STATE_LOCKED instead of hardcoding its value.
- In IRP_MN_VERIFY_VOLUME handling, lock the volume before playing
with it, and again let the volume be dismounted only if OpenCount == 0
(and the IoDeleteDevice() call is done by RawCheckForDismount()).
This will allow compatibility with modern OSes and
modern disk management utilities.
It will also improve performances by properly aligning
partition start.
And it will let enough room at the begin of the disk
for 3rd party bootloaders.
WARNING: this is not compatible with previous partition
model, and old one will likely not be compatible. You'll
have to erase your whole partition table and start from
scratch.
Simplified implementation of the proposed solution by George Bișoc
and revised by Victor Perevertkin from PR #952.
NOTE: For whatever reason ERROR_INSTALL_BOOTCODE and ERROR_WRITE_BOOT
seem to be redundant with each other. To be investigated...
Most of Linux filesystems have the MBR ID assigned as 0x83 and when the user chooses BTRFS upon partitions page wizard, we're thinking of the partition as having Ext2 filesystem which is wrong and misleading.
- Interface the TreeList code and populate it with the list of
discovered disks and partitions. Code is adapted from USETUP.
Also display the names of the disks.
- Display some installation settings summary, before doing the
installation proper.
- Force the user to select a checkbox when (s)he acknowledges that
ReactOS is alpha-quality software and may break on his/her computer
or corrupt his/her data.
- Improve wizard pages transitions and buttons enabling/disabling.
- Press Shift-F10 to start a command-line (as in the 2nd-stage setup
in syssetup.dll).
- Use some explicit UNICODE functions/macros.
Add the TreeList control from Anton Zechner and Sébastien Kirche
from https://github.com/sebkirche/treelist (under GPL-3.0+) as a
*TEMPORARY* solution until a better-suited control for ReactOS is
developed.
- Compilation fixes for the TreeList control.
- SETUPLIB and USETUP: Convert PARTENTRY::DriveLetter to WCHAR.
- SETUPLIB: Retrieve volume label.
- SETUPLIB and USETUP: Move the partition types (IDs) table from
USETUP to SETUPLIB so that they can share it with the 1st-stage
GUI installer too. Also take the opportunity to update the table
with information from http://www.win.tue.nl/~aeb/partitions/partition_types-1.html
and cross-checked with the list from Paragon Hard-Disk Manager.
- USETUP: Simplify PrintPartitionData().
- Add some code comments.
This increases performance for each time the SETUPLIB calls (using NT paths)
Win32 SetupAPI functions which of course only accept Win32 paths.
- Handle also the fact that a NT path to convert may start with
\Device\HarddiskX\PartitionY\..., which can be a symlink to
\Device\HarddiskVolumeN\... on some systems. In that case, the
Win32 path mapping should be done slightly differently.
- Add support for network mapped drives.
- Improve UI usage and wizard transitions (Work In Progress).
- Add interfacing code that display installation status.
- When the user attempts to cancel the installation during file copying,
pause the file copying, and restart it if the user changes it mind and
wants to continue the installation. Otherwise file copying is properly
halted.
- BOOTDATA: Use standard INF signature string, so that they can be
opened successfully using ReactOS' or Windows' setupapi.dll with
the INF_STYLE_WIN4 style.
- SETUPLIB: Use the correct INF_STYLE_* INF styles in SpInfOpenInfFile() calls.
- REACTOS : Switch thread locale to user-specified LocaleId when calling
SetupOpenInfFileW(), so that the correct localized strings are used.
- In PrepareCopyInfFile(), explicitly use AddSectionToCopyQueueCab()
to retrieve the files from the CAB INF.
Otherwise if one continued to use and rely on AddSectionToCopyQueue()
to do this job instead, bad things could happen like, re-enumerating
twice or more (and adding them to the file queue) the same files.
This is because in the new code AddSectionToCopyQueue() can be called
many times to deal with different SIF file sections. The old code then
called AddSectionToCopyQueueCab() in turn, but on the same list...
Now the steps performed are clear, and everything works as intended.
- Simplify the code that computes the full source and destination paths
for the files to be copied and directories to be created, using helper
functions.
- Compute the source file and target location from the TXTSETUP.SIF
entries in a NT-compatible manner.
- Define IntUnicodeStringToBuffer function to copy the contents of a UNICODE_STRING to a string buffer with a UNICODE_NULL correctly.
- Use it.
JIRA issue: N/A
- Use adequate access flag when opening symbolic links.
- Simplify the prototype of UpdateRegistry() since now both Setup INF
handle and settings lists are inside the USETUP_DATA structure.
This means, call in the correct order the user callback with the correct
parameters (in particular the correct paths for file copy operations),
and check also for the callback returned value to know whether or not to
continue the file operations.
This allows using some of the SetupApi.dll functions when SETUPLIB is
used in the (Win32) GUI 1st-stage installer "REACTOS", while using the
custom implemented NT-aware functions in "USETUP".
- Add support for delete and move/rename operations, which are needed
for implementing ReactOS upgrading support.
- Use cabinet contexts.
- Use standard LIST_ENTRY structures for implementing the lists.
- Move the path-building hack code in SetupCommitFileQueueW() that had
been introduced in r66604 (97bb83f) out of the file-queues code.
- Make the function prototypes compatible with win32's setupapi functions.
- Fix the format of the data passed to the custom notification handler.
- Adjust the file-copy callback to correctly use its arguments (setupapi-compatible).
Old cursor looks better.
We revert to SVN state r75865 and note the diff for Amine
so that the wrong version won't come back in the future.
Old file is just 326bytes while the one that we do not want is 33.134bytes
which means we will save some RAM as well.
- Move a great deal of global variables into the USETUP_DATA structure
(the SetupInf, the SetupFileQueue, the generic lists...).
- Place the common setup initialization code into an InitializeSetup()
routine, and the cleanup code into FinishSetup().
- Implement the setup-code part support for the TXTSETUP.SIF setup
source path override variables "SetupSourceDevice" and "SetupSourcePath"
(see CORE-9023); support for them in SETUPLDR will be added later.
Now, we make sure that we update ref count and BCB list membership
with the BCB lock held, in a row.
This will avoid race conditions where the BCB was removed from the
list, then referenced again, leading to inconsistencies in memory
and crashes later on.
This could notably be triggered while building ReactOS on ReactOS
(one would call this a regression).
CORE-15235
- In the 1st-stage GUI setup, display the list of existing discovered
NTOS installations, with its name, installation path and vendor name,
and an associated icon.
svn path=/branches/setup_improvements/; revision=75544
[SETUP:REACTOS] More interfacing of the 1st-stage GUI installer with the setuplib.
- Add a ConvertNtPathToWin32Path() helper function that allows converting
NT paths like: \Device\CdRom0\i386\txtsetup.sif file into a Win32 path
that the Win32 versions of the SetupApi functions (e.g. SetupOpenInfFileW)
can accept.
Note that RtlNtPathNameToDosPathName() cannot be used here because this
function only works for NT paths within the \DosDevices\ (or \??\) NTObj
directory, and not in other directories.
- Use this helper function in the implementation of SetupOpenInfFileExW.
Now the txtsetup.sif function can be loaded by setuplib *when being used*
by the 1st-stage GUI installer!
- Remove deprecated code.
svn path=/branches/setup_improvements/; revision=75668
svn path=/branches/setup_improvements/; revision=75721
svn path=/branches/setup_improvements/; revision=75750
- Remove a non-recurring job from the job list after starting it.
- Remove a recurring job from the start list, calculate its next start time and insert it again.
- Calculate the timeout for the next job.
This includes also to remove the created symlinks.
Symlinks deletion is special, as one has to open first the symlink
itself (and not its target), then remove the "SymbolicLinkTarget" value
before really deleting the key. Of course everything must be done under
proper access rights.
Additional changes:
- Change prototype BOOLEAN CmpLinkKeyToHive(...) to NTSTATUS CreateSymLinkKey(...).
- Silence few DPRINTs.
- Document some FIXMEs that I need to inspect later on ReactOS.
- HKEY --> HANDLE.
- It can therefore be also used by the 1st-stage GUI setup.
- Rename some function parameters to clarify what they should be.
Based on:
svn path=/branches/setup_improvements/; revision=75750
Additions:
- Use explicit __cdecl calling convention for variadic functions.
- Fix also the whitespace in consup.h.
svn path=/branches/setup_improvements/; revision=75749
[USETUP] Improve usage of ERROR_TXTSETUP_SECTION and ERROR_CABINET_SECTION error codes, correctly display associated error message, and remove a deprecated string.
- Apart from allowing a UI cache variable that may be used when
displaying GENERIC_LIST_ENTRY-ies, do not store any display strings
associated to these list entries. They should be instead computed only
when initializing a list UI (or a combo-box or list control if the
code is used in Win32 environment).
For this matter a callback is provided to InitGenericListUi() that
does the job of computing the displayed string corresponding to a
given GENERIC_LIST_ENTRY.
- Simplify the calls to InitGenericListUi(), and refactor the
RestoreGenericListUiState() function.
- Use for-loops for iterating over GENERIC_LIST items.
- Adapt the storage data format for lists of settings items.
- The txtsetup.sif INF format specified in LoadSetupInf() should not be
INF_STYLE_WIN4 (to be investigated...).
- Use OBJ_CASE_INSENSITIVE when initializing object attributes
(no actual reason why to keep case sensitivity there).
- Check the success of a RtlStringCchPrintfW call in EnumerateReactOSEntries().
- Explicitly check for returned STATUS_NOT_SUPPORTED from ChkdskPartition()
or FormatPartition(), and display an appropriate error message.
- Remove some left-over comments but also explain why I kept some
commented code (mainly for future reference).
Comment about the status of two '#if 0'.
[SETUPLIB] Fix some uninitialized variables (caught by clang-cl).
Also, initialize NtName->Length in return paths in ResolveArcNameManually().
Only convert to ANSI when needed (e.g. in the display code for usetup).
The 1st-stage GUI setup will however use the UNICODE strings directly.
svn path=/branches/setup_improvements/; revision=75753
[SETUPLIB] Copy the mui* files from usetup into the setuplib for later usage, and add mui.c to build.
svn path=/branches/setup_improvements/; revision=75711
[SETUPLIB][USETUP] Start massaging the MUI files and split up what concerns only the usetup "resources",
and what concerns general language parameters (in the setuplib).
It may be interesting to retrieve the language parameters
from INF file (e.g. intl.inf) (suggested by Giannis).
svn path=/branches/setup_improvements/; revision=75715
[SETUPLIB] Add the necessary headers to mui.c to be able to compile.
[USETUP] Comment out the languages that don't have any resources.
svn path=/branches/setup_improvements/; revision=75716
[SETUPLIB][USETUP] Adjust MUI and settings code.
- Re-enable settings code concerning language, keyboards & layout
in the setuplib, and remove the old code in usetup.
- Remove useless code in setuplib's mui.c.
- Rename usetup's MUI "LanguageList" into "ResourceList" so as to
avoid colliding with the "LanguageList" of setuplib.
- Add the magic CMakeLists line "add_definitions(${I18N_DEFS})" that
I forgot previously, that caused the "LanguageList" of setuplib to be empty!
The code compiles and works during my tests.
svn path=/branches/setup_improvements/; revision=75717
[SETUPLIB] Change some string pointer types & function prototypes so that the usetup-specific
"SelectedLanguageId" variable is not used in the library.
svn path=/branches/setup_improvements/; revision=75719
[SETUPLIB] Change some function prototypes so that the usetup-specific "SelectedLanguageId" variable is not used in the library.
Also, make AddEntriesFromInfSection(), DefaultProcessEntry() and the typedef PPROCESS_ENTRY_ROUTINE private again.
svn path=/branches/setup_improvements/; revision=75720
svn path=/branches/setup_improvements/; revision=75724
[USETUP] Code adaptations.
- Transform some do{}while() loops into while(){} ones, since the lists on which we work may only contain one single {NULL} element.
- Modify MUIGetOEMCodePage() call after r75719.
- Use PCWSTR where needed.
svn path=/branches/setup_improvements/; revision=75722
[USETUP] Adjust some MUI & settings calls, following the commits r75719, r75720 and r75721.
svn path=/branches/setup_improvements/; revision=75723
[INPUT.CPL] Adjust inclusion of "muifonts.h" (commit 3dbd44f) due to changes introduced in r75711 and r75715.
Use WaitForMultipleObjects in the mail scheduler loop:
- Use events to signal service stop and job update events to the main loop.
- Use the timeout timer to start the next job.
Preparation to support pen width.
Split IntGdiWidenPath function from PATH_WidenPath function.
IntGdiWidenPath is independent from DC.
CORE-2527, CORE-13534
This code is moved ther since it'll be needed for 1st-stage GUI
installer as well.
The code dealing with keyboard layouts is however not moved yet,
since it uses MUI* functions from usetup.
I'll try to think how to include that in setuplib...
svn path=/branches/setup_improvements/; revision=75701
Move the bootsup.c code into the setuplib, as the code is generic
enough to be used by both USETUP and the 1st-stage GUI installer.
svn path=/branches/setup_improvements/; revision=75674
- Implement user32!SwitchToThisWindow by NtUserCallTwoParam TWOPARAM_ROUTINE_SWITCHTOTHISWINDOW.
- Improve user32!CloseWindow with using SetActiveWindow and ShowWindow (synchronized).
CORE-15165
This helper is inspired by the reactos/reactos.c!LoadGenEntry() function.
Use this helper in the different Create*List() functions.
svn path=/branches/setup_improvements/; revision=75700
- Compute the installation source paths based on the full path of the
installer program that uses the setup library.
- Add INF_STYLE_OLDNT define in infsupp.h.
- Add some (silenced) diagnostic DPRINTs.
svn path=/branches/setup_improvements/; revision=75667
- Use correct inf style flags in SetupOpenInfFileEx() calls when opening
txtsetup.sif and unattend.inf. Technically txtsetup.sif would be
INF_STYLE_WIN4, but since we use "$ReactOS$" as its version signature,
it would not work when opening it with setupapi.dll functions.
Hence this flag is combined with INF_STYLE_OLDNT too.
- Don't fail if opening the \SystemRoot symbolic link doesn't work
(usually due to incorrect access rights); in that case, just use the
installer image file path as the installation source path.
svn path=/branches/setup_improvements/; revision=75676
And turn the previous two functions into macros.
svn path=/branches/setup_improvements/; revision=75635
svn path=/branches/setup_improvements/; revision=75652
Used in exiting code and in later code in this file.
svn path=/branches/setup_improvements/; revision=75632
svn path=/branches/setup_improvements/; revision=75652
Following r75518, we use the $winnt$.inf file (created in System32
by the 1st-stage installer) as the setup information file for the
2nd-stage setup for:
- retrieving the installation source media path;
- retrieving the unattended information that was copied from the
unattend.inf file from the installation source media.
The installation source media path is converted from NT format to
Win32 format for usage with Win32 functions and storage in the registry:
this is done by GetInstallSourceWin32(), which replaces the hackish
GetRosInstallCD() function.
The $winnt$.inf file is also updated, and the registry values "SourcePath"
and "ServicePackSourcePath" are created / updated in:
HKLM\Software\Microsoft\Windows\CurrentVersion\Setup .
svn path=/branches/setup_improvements/; revision=75524
- Move several global setup variables into a structure "USETUP_DATA",
similar to the syssetup structure "SETUPDATA" (or the WIP 1st-stage
installer structure of the same name), so that these variables can be
set easily by different helper setup functions;
- Move CheckUnattendedSetup() and GetSourcePaths() to setuplib and make
CheckUnattendedSetup() use the USETUP_DATA structure;
- Add a LoadSetupInf() function that loads the txtsetup.sif file
(factoring out the corresponding code in USETUP);
- Add a InstallSetupInfFile() function (that I'll probably rename later on)
whose purpose is to create a valid "$winnt$.inf" setup installation file
in the ReactOS\system32 directory, which should help the 2nd-stage installer
to correctly retrieve the source installation media we used during 1st-stage,
and contain the unattended setup lines copied from unattend.inf. This is
done in a Windows-compatible way.
svn path=/branches/setup_improvements/; revision=75518
[USETUP] Close the txtsetup.sif file at the end of the operations.
svn path=/branches/setup_improvements/; revision=75539
In windows the toolbar itself creates the search shell extension. In ros we send a command to the CShellBrowser to let it handle it itself. Use the same command for showing and hiding it.
CBaseBarSite: Ask the CBaseBar to close itself when the x button is pressed.
CBaseBar: Hide the bar and inform its site that it is closing when it gets the close command.
CShellBrowser: Cache the guid of the current vertical bar and use it to report correct command status to the toolbar.
Also implement toggling the Folders, Favorites, History and search commands.
CInternetToolbar: Query the Folders and Search command status from the site so that they can be properly be shown as pressed.
In Windows, user32!CloseWindow minimizes the window. It doesn't close the window actually. This is not a joke. This bad function naming is responsible to MS.
They currently do nothing, since the getter functions don't actually
capture (copy) the strings but merely return pointers to read-only strings.
But the calls are placed here for consistency, because if one day the
getters' implementation is changed so that strings are captured, it would
then be needed to free the allocated buffers.
In addition, fix a buggy call to INF_GetData() -- should be instead
INF_GetDataField() -- in AddSectionToCopyQueue().
svn path=/branches/setup_improvements/; revision=75516
Remove the commented-out IsValidPath() in that file, and remove
as well the temporary prototypes in regutil.c .
svn path=/branches/setup_improvements/; revision=75249
Split the registry helper code into registry utility functions
(create/mount/unmount/verify hives) and registry creation code
(the rest: initializing the system hives) and move it in the
setup library.
svn path=/branches/setup_improvements/; revision=75247
[USETUP][SETUPLIB] Improve how the Setup* INF APIs are interfaced in the code.
Define prototypes compatible (including their calling convention)
with the ones defined by setupapi (.h/.dll) so that it can be possible
to either use an internal implementation of these functions (via the
INFLIB library) as currently being done, or using imported setupapi
functions, as would be done by the future 1st-stage GUI installer.
[SETUPLIB] Cleanup the registry.c file and use the infsupp.h header.
svn path=/branches/setup_improvements/; revision=75345
svn path=/branches/setup_improvements/; revision=75346
[USETUP][SETUPLIB] Move some INF-related code from usetup to the setuplib.
- Move the generic INF_GetDataField() and INF_GetData() helpers to
setuplib, and rework them a bit so that they explicitly call setupapi
functions (or implementations thereof when being used in usetup);
- Rework the headers in accordance;
- Fix compilation in lib/registry.c .
- Fix compilation when these headers are used withing usetup (who
doesn't use setupapi.dll) and "reactos" (the 1st-stage GUI installer
that uses setupapi.dll).
svn path=/branches/setup_improvements/; revision=75515
svn path=/branches/setup_improvements/; revision=75537
svn path=/branches/setup_improvements/; revision=75538
- As GetSourcePaths() is used once in USETUP to initialize global
UNICODE_STRING path strings once, move it out of drivesup.c and
put it in usetup.c. Then remove drivesup.c : 1 file less!
- Move some INF file prototype declarations out of usetup.h and
inside inffile.h where they should better be, as inffile.h and .c
is the glue code for the INF library, defining similar functions
as the ones in setupapi.dll.
- I rename our local SetupOpenInfFileW() into SetupOpenInfFileExW()
because the latter one takes an extra user-provided LCID parameter,
and this is this one that we use in USETUP.
- Make 'UNICODE_STRING SourcePath;' visible only inside usetup.c
(not used elsewhere).
- Implement installation path validity check in case we are either
in repair/update, or unattended setup mode. If the path is detected
as invalid, then we fall back into manual path specification
(for now...; note that we could instead fail the installation too).
svn path=/branches/setup_improvements/; revision=75246
- Modify a bit the page flow so that the upgrade page is inserted before
the Device-settings page, and after the Install-Intro page.
- Insert some extra 'RepairUpdateFlag' checks in SelectPartitionPage() and
InstallDirectoryPage() to take specific actions in case the flag is TRUE.
svn path=/branches/setup_improvements/; revision=74576
svn path=/branches/setup_improvements/; revision=75525
This fix is needed for the Delphi GUI programs.
- Don't enumerate the windows whose owner window exists.
- Use PostMessage SC_RESTORE instead of ShowWindowAsync SW_RESTORE.
CORE-15165
Implement work-in-progress code that, when upgrading an existing
ReactOS installation, verifies whether the (existing) system registry
hives are valid (by loading & unloading them, this allows using the
built-in repair functionality if needed), or not.
If a given hive cannot be repaired successfully, it is backed up
(with a '.brk' extension, "brk" as "broken"), then is marked up for
recreation.
When all hives have been checked, if there are any hive that needs
to be recreated, we recreate its hive file, mount it, but we do *NOT*
mount the other valid existing hives for update. We create instead
dummy registry hives so that we can actually use, as the update code,
the same one as the one used when fully creating the registry hives
for a clean installation (of course, this choice can be improved later on).
The update code (i.e. the same as the registry clean-install one) then
adds the registry keys & values, either putting them in the dummy
registry hives (the ones that we don't want to recreate) or in the
registry hive that is recreated.
At the end, the (re)created registry hives are flushed back to disk,
and a copy of them (under a '.sav' extension) are created, so that they
can be used for restoration purposes if 2nd-stage (and up) goes berserk.
Extra fixes:
- Use the correct structure member field when initializing the
'InstallDir' variable, when performing an upgrade.
- CreateNestedKey() should be better analysed to see whether it correctly
creates the full registry path compatible with volatile/non-volatile keys
(under inspection).
svn path=/branches/setup_improvements/; revision=75010
Will be used soon to make backups of system files, like the registry hive files just freshly created.
- Make the SetupCopyFile() function closer to its win32 counterpart.
- Adjust the code that calls SetupCopyFile().
svn path=/branches/setup_improvements/; revision=75008
svn path=/branches/setup_improvements/; revision=75009
The rationale is as follows.
We now have the 1st-stage ReactOS setup running with its own registry SYSTEM hive,
similarly to regular ROS running instances (livecd, regular installation...).
The ReactOS-specific SetInstallPathValue() hack, introduced in a76689e9 (r3794)
and 5f973ce0 (r3795), is removed. This hack told the kernel that, during the setup,
it had to "switch" the used registry hives and instead use the ones of the ROS
installation being prepared.
This was really hackish because this means, mixing between registry settings used only
for the setup running instance, that could use different registry settings
than the ones that should be set for the ROS installation being actually performed.
Also, note that in the case of a 1st-stage GUI setup, consisting in running the
LiveCD + the GUI setup program, this situation would be untenable. Note also that
for people willing to use the Setup*** functions exported by setupapi.dll to parse
the registry INF files to initialize the registry of the ROS installation being
prepared, this would be impossible either.
Hence the need to have offline registry modification functionality.
svn path=/branches/setup_improvements/; revision=74766
We now run the 1st-stage setup with a regular system hive, similarly to
what's done for the LiveCD, or for a regular ROS installation.
The ExpInTextModeSetup hacks I previously removed are now completely unneeded.
svn path=/branches/setup_improvements/; revision=74762
Split the "main" winldr.h header into the one containing global code
that actually doesn't really depend on the "windows" NT loader part,
and one that actually concerns code just for the NT Loader.
The latter goes into "ntldr".
svn path=/branches/setup_improvements/; revision=74743
- Rework CmpSetSystemValues() and remove its 1st-stage text-mode setup hack, since a real registry hive will be used for 1st-stage either.
- Lock, then unlock the registry in NtInitializeRegistry when initializing the hives & flusher.
- Call CmpInitializeHiveList() (i.e., initialize the other hives like \Software, \User, \.Default) only when we are not in setup-mode.
svn path=/branches/setup_improvements/; revision=74747
This reverts a7c26408 (r53255) and ff75ae1b (r53694), and a hack from 6075ae9a (r46690).
svn path=/branches/setup_improvements/; revision=74745
svn path=/branches/setup_improvements/; revision=74746
Add a mechanism that detects via the version resource whether it's needed to mirror
the application layout or not. This is a "follow up" to commit 5b14b6f.
This code is taken from the GetProcessDefaultLayout() function from Wine
(called there "WIN_CreateWindowEx").
In Windows the version-resource mirroring happens when the GDI is loading the LPK.
- Implement mirroring via version resource. GDI linking on startup is required.
- GDI32: Add LPK loading on GDI startup.
To enable mirroring via version resource the lpk.dll must be loaded before any windows are created.
This is done through GdiInitializeLanguagePack() function which is called in GdiProcessSetup().
- LPK: Use Wine debug messaging.
- Also, reserve a new field to display whether the disk is MBR or GPT-initialized.
- Adjust the code that used the "STRING_HDDINFOUNKx" string labels.
The extra slot added to some of the strings is to indicate whether the disk is MBR or GPT.
As attentive readers will notice, the code that "determines" whether the disk is either MBR or GPT
is of course just temporary: it's just a placeholder code that should currently just show "MBR".
Having the NoMbr flag mostly signifies that the disk is uninitialized. A GPT disk also has a MBR
but it's for "backcompatibility" / "protective" reasons (so that MBR-only tools would see the disk
as just a big 1-partitioned only disk with which they could barely do anything). A real involved test
for the nature of the disk should involve calling the correct IOCTLs in the disk detection code in partlist.c .
This is for the future (and when ReactOS will better support GPT disks).
svn path=/branches/setup_improvements/; revision=74614
svn path=/branches/setup_improvements/; revision=74615
- The "intro" page is renamed into the "Welcome" page, because its corresponding resource indeed is the welcome screen;
- Because the "setup start" page can only be displayed once, move it out of the while-loop, and use its result as the initial value of the 'Page' variable.
- Remove unneeded _PAGE_NUMBER_DEFINED guards;
- Add a DPRINT in the RepairUpdateFlag case of RegistryPage() (because we don't implement yet a correct upgrading or repairing of the registry).
- In InstallIntroPage(), display the page itself only if needed (i.e. after all the validation checks & repair/update or unattended checks are done). Similar modifications are done also in DeviceSettingsPage(), SelectPartitionPage()
- Turn both CheckUnattendedSetup() and UpdateKBLayout() into static functions.
- Fix the code of ScsiControllerPage() so that it can be compiled if needed, and add a dummy OemDriverPage().
svn path=/branches/setup_improvements/; revision=74575
- Improve the capture of OBJECT_ATTRIBUTES parameters that are passed
(by pointer) to the Cm* helper functions, and the capture of
UNICODE_STRINGs.
- Correctly differentiate user-mode vs. kernel-mode root directory handles
(in OBJECT_ATTRIBUTES): note that most of the Cm* APIs expect their
parameters to be kernel-mode (pointers, handles...).
CORE-13448
- Make the strings translatable.
- Include the English translation within the main resource file.
- Use ConResPrintf() and ConResPuts() for resource strings.
- Validate the information class parameter in NtQueryValueKey().
- Call the post-callback in NtSetValueKey() only if the callback
has been registered and the CmSetValueKey() call is executed.
Some applications, run on a CDROM, may attempt to open
a file with write access. CDFS driver will deny such
request in ReactOS (but also in Windows NT). Then, to
restore that behavior from old Windows (9X), our ntvdm,
as Microsoft ntvdm will attempt to reopen the file only
with read access, if opening failed because of denied access
and if the file is on a CDROM.
CORE-15211
Fixes in particular a regression brought by 57402ee (thanks George Bisoc
for having noticed this), where the BTRFS bootsector code could not be
installed anymore.
CORE-15192
Fixes crashes when the tool is compiled with Run-Time Checks enabled.
And thus should fix the MSVC buildbots!
Thanks to Pierre for having brought this issue to me.
CORE-15201
- Fix the usage of the BootEntry's "Version" member.
- Don't surround with too many quotation marks the "friendly" boot entry name in AddBootStoreEntry().
svn path=/branches/setup_improvements/; revision=74964
In particular the helper functions CreateCommonFreeLdrSections() and (Un)protectBootIni() are now removed from there (they are used in bldrsup.c only).
This should pave the way for future integration with other sorts of NT boot loaders (BootMgr and (u)EFI boot loader).
svn path=/branches/setup_improvements/; revision=74954
This concerns so far, only freeldr.ini and to an extent, boot.ini, but planning in the future to add support for registry-oriented BCD
as well as possibly direct (u)EFI entries, using the corresponding NT functions.
This code is so far only used in osdetect.c, but will be soon used by usetup's bootsup.c (note that some helper functions: CreateCommonFreeLdrSections()
and (Un)protectBootIni() are already taken from it and used in bldrsup.c).
- In EnumerateNTOSBootEntries(), continue enumerating the boot entries until the user callback returns an status code that is not successful.
- Remove some old code from osdetect.c; use directly BootEntry->FriendlyName when building the display names of the available installations, since now
BootEntry->FriendlyName is a PCWSTR (and not a UNICODE_STRING anymore).
svn path=/branches/setup_improvements/; revision=74943
[SETUPLIB] Make the NTOS_BOOT_ENTRY structure more generic, so that it can wrap around either actual NTOS boot entry options, or FreeLdr-like boot-sector options.
In a sense, the NTOS_BOOT_ENTRY structure now looks much more like the NT structure "BOOT_ENTRY".
- Adapt the code in bldrsup.c to these modifications, and re-enable FreeLdr-like boot-sector-file support code that was commented out.
More code cleanup will follow later.
svn path=/branches/setup_improvements/; revision=74952
- Convert almost all swprintf() into StringCchPrintfW() and wcscpy() into StringCchCopyW();
- Explicitly add a trailing path separator to the "\Device\HarddiskX\PartitionY(\)" paths
when they refer to FS directories (and not to partition objects);
- Remove useless (and half-buggy) "Remove trailing backslash" code.
With that, it is possible to install ReactOS in e.g. C:\ReactOS (as usual), C:\ReactOS\dir1\dir2 (as many dirs as you wish), and also in C:\ (yes yes!).
But in that latter case, a strange bug related to the registry arises...
Additionally:
- Adjust some comments;
- Add some debugging DPRINTs;
- The SetInstallPathValue() is part of the big hack I've mentioned in f51faa4a (r74709).
svn path=/branches/setup_improvements/; revision=74717
Also, trim any potential trailing path separator from the NT disk partition path before opening the partition object itself for read/write access.
svn path=/branches/setup_improvements/; revision=74715
- Fix a bug in RegpCreateOrOpenKey() when we attempt to open an
non-existing (sub-)key.
- Check failure of RegCreateKeyExW() in CreateSymLink().
- Remove unneeded includes.
- Update few comments.
We won't reuse a BCB created for mapping, we will now have
our own dedicated BCB.
This allows having a bit more cleaner implementation of CcPinMappedData()
We now handle race conditions when creating BCB to avoid
having duplicated BCB per shared maps.
Also, we already specify whether the memory will be pinned
when creating the BCB, to avoid potential duplications or
BCB misuse.
This fixes optional modules being installed in places
that make no sense.
This also fixes auto installation of Gecko. Thus, this
fixes unattended setup of ReactOS for testbots.
CORE-15160
- Fix parsing of the options.
- Only uppercase the file name part and NOT the full path!
And do it in a way GCC-Linux correctly understands, aka.:
*ptr = toupper(*ptr); ++ptr;
but NOT!:
*ptr++ = toupper(*ptr);
(that last one worked on GCC-Win and MSVC).
[CMAKE] Let's keep SETUPREG.HIV and BCD hive file names in uppercase
(use the '-u' switch) while keeping the other ones in lowercase.
Should definitively fix GCCLin builder!
- Add support for '/' switch characters.
- Add '-?' switch.
TEMPORARY:
- Add a '-u' switch to generate hive file names in uppercase (by default
we keep these in lowercase). Should help in fixing GCCLin builder.
[CMAKE] Fix expected file name case. Should help in fixing GCCLin builder.
This commit is needed for building a single bootcd registry hive. It will also enable 1-st stage setup to have a proper registry present as done on Windows.
CORE-13347
[CMAKE]
Note that the mkhive commands should really depend on the generated UTF16 INF files since the latter are those that are actually used as input for mkhive.
Otherwise, the mkhive calls & the UTF16 INF file conversion is not serialized and we can generate "corrupted" hives due to the fact that mkhive is using
INF files that are in the process of being (and therefore, only partially) generated.
svn path=/branches/setup_improvements/; revision=74741
svn path=/branches/setup_improvements/; revision=74749
- Transform TryToStartShell() into a StartProcess() function whose aim
is just to start processes, since there is no extra initialization needed
for starting a shell process.
- Modify StartInstaller() to call StartProcess() for starting the ReactOS GUI
installer found on the installation media, from a path automatically expanded
depending on the ambient CPU architecture.
svn path=/branches/setup_improvements/; revision=75331
- Don't hardcode the installer file name in ExpandInstallerPath() but
provide it from a parameter. Return TRUE/FALSE if the installer file
is/is not found.
- Expand any environment string present in the items command line.
- Enable/disable the menu buttons depending on whether or not an
associated command is present.
- Enable mirroring from resources.
Add a test program that shows how the *PSMTextOut() functions draw
the strings by a side by side comparison between LpkPSMTextOut() and
UserLpkPSMTextOut().
[LPK][USER32] Fix reading layout checks.
- Both functions don't care about DT_RTLREADING flag.
- Added a different method to check reading order.
[USER32] Remove gcc-build breaker semicolon.
This function is a condensed version of LpkPSMTextOut() (from lpk.dll),
but not bug-for-bug equivalent.
The main difference is that the underscore is drawn without taking into
account any kind of script processing.
- The function is almost fully implemented, but some return values change wildly.
The results are mostly compatible to the Windows results.
- Remove #ifdef (this is primary ReactOS code).
- Add a pointer to the current parameter.
- Change the 'present' and 'not present' radio buttons to auto radio buttons.
- Display present and non-present optional parameters properly.
If so, return such BCB instead of creating a new one. This will
allow (at some point) to be more consistent in case of concurrent
mapping.
This fixes a few CcMapData tests.
Given current ReactOS implementation, a VACB can be pinned
several times, with different BCB. In next commits, a single
BCB will be able to be pinned several times. That would
lead to severe inconsistencies in counting and thus corruption.
This quick check based on bits and operation is for 2^ based
sector sizes (most of the cases) and will perform faster than
the modulo operation which is still used in fallback in case
the sector size wouldn't be a power of 2.
Because of FontForge's bug, the underline position value of Tahoma and Tahoma Bold was wrongly set (#789, #809). This commit fixes those values.
CORE-15079
- Deal with expanded paths only when loading icons. If the user specifies
an unexpanded path to a module containing icons, this unexpanded path
is returned to the caller!
- Perform the dialog cleanup in WM_DESTROY message handler.
- When pressing ENTER/OK in the dialog box, retrieve the selected icon and
close the dialog *ONLY WHEN* the user actually has selected an icon.
If the user modified manually the module path, pressing ENTER/OK instead
reloads the icons list.
This leaked at least one IRP for every write, making it the largest leak when
running with BTRFS as the system volume.
Thanks to Victor Perevertkin.
Submitted upstream thanks to Pierre:
https://github.com/maharmstone/btrfs/pull/106
IopCloseFile can be called by IopDeleteFile. In that situation, it
doesn't set any process as first parameter. Furthermore, we are in a
situation where it's not required to lock the file object (see the
assert before the call).
- Flush the new control set after the control set has been created.
- Make sure that we create a new control set only on the first boot after setup.
- Move the control set copy code to a separate function.
- Get rid of the global control set values.
- Check whether the user-provided token is a primary token.
- Do not fail when the RtlAdjustPrivilege() call fails (see the code
comment for an explanation). TL;DR is: that call may indeed fail but
the privilege may also not be necessary because the user-provided
token is a restricted version of the caller's primary token.
And this is situation is perfectly fine.
This fixes Java 7 installation, CORE-14874.
- SeIsTokenChild(): Correctly check whether a caller-provided token
is a child from the current process' primary token by looking at
its ParentTokenId member.
- Add a SeIsTokenSibling() helper to determine whether a caller-provided
token and the current process' primary token are siblings, by comparing
their ParentTokenId's and AuthenticationId's.
NOTE: Children tokens are created through CreateRestrictedToken();
sibling tokens are created through DuplicateToken() (amongst others).
See slide 49 of https://www.slideshare.net/Shakacon/social-engineering-the-windows-kernel-by-james-forshaw
or https://googleprojectzero.blogspot.com/2016/01/raising-dead.html
for some details.
Now that SetupDiCreateDevRegKeyW is working as expected, we can no
longer (ab)use it to retrieve a ports resources. Use SetupDiGetDeviceInstanceIdW instead and open the registry keys manually.
Also comment about the observed behaviour of the returned number of "characters"
returned by the ANSI versions of these APIs (which is tested by advapi32_winetest:service).
[ADVAPI32:SCM] Add a comment concerning wide characters vs. bytes mismatch.
This allows the code to give a last chance for expanding the string.
And if it still fails, we bail out as expected and the caller will
take appropriate actions.
As a by-product, this commit (together with the previous one e2960914)
fixes the regression CORE-14981 (Quick-Launch bar couldn't be opened
in LiveCD).
This could be triggered when attempting to read/write to really big
files. It was causing an attempt to read 0 bytes in Cc, leading to
asserts failure in the kernel (and corrupted file).
CORE-15067
Fixes accessing nullptr / invalid handle in Window->SystemMenu
I could trigger this occasionally when moving Opera 12.18 window
in front of Spotify 1.0.29.92 window.
Fix was:
authored by Thomas Faber (Thank you!)
tested by me
reviewed by James Tabor
The Logoff-Handler might try to notify the scheduler service which might already have been shut down. We need to fix the shutdown sequence, so that the service manager stops all running services.
Might fix CORE-14521.
- Encrypt passwords before passing them to the service manager. Right now, we are using a fixed encryption key. This will be fixed later.
- Replace the calls to ZeroMemory which are used to wipe the password buffers by calls to SecureZeroMemory.
These tests show that the required output buffer size it returned in out.Length field when STATUS_BUFFER_TOO_SMALL is returned.
This patch will be sent upstream.
This will also fix MS Office 2010 installation, where the installation
expects the NetworkService account profile to be already loaded (this
condition is always verified on Windows since RPCSS and other services
also run under this account).
CORE-14988, CORE-14989
If we cannot find the functions in kernel32, try to get them from
kernel32_vista. This will allow testing our own implementation, while
still failing in W2K3 (no implementation available).
This avoids keeping connections open and reopening them as soon
as the user reopens their session.
This fixes network shared being duplicated each time session
is reopened.
CORE-15012
LibreOffice started to show GUI in a mono-spaced font after 0.4.10-dev-466-g
35f62fc5ba
We can fix this by partial revert
as proposed by Katayama Hirofumi MZ, who was author of those changes as well.
We revert the changes in penalty system here.
Add code that demonstrates the use of the lpOrder and lpCaretPos struct members of GCP_RESULTS w/ and w/o GCP_REORDER flag.
Changed demo string literals to arrays containing character unicode values.
- Try to make use of ScriptStringCPtoX when trying to fill the lpCaretPos struct member;
if USP10 error happens fallback to the unreorder method.
- Remove now incorrect debug print.
- Add the MUI ID resource.
- Implement the shutdown timeout progress bar function and the TimeOut string handler.
- Add the resource string to other resource languages.
- Use the registry value AdminComment for admin comments instead of Description. Fix SamrSetInformationGroup as well.
- Add some TRACEs.
- Do not fail if the Members value does not exist.
Currently, our CcMapData() behavior (same goes for CcPinRead()) is broken
and is the total opposite of what Windows kernel does. By default, the later
will let you map a view in memory without even attempting to bring its
data in memory. On first access, there will be a fault and memory will
be read from the hardware and brought to memory. If you want to force read
on mapping/pinning, you have to set the MAP_NO_READ (or PIN_NO_READ) flag
where kernel will fault on your behalf (hence the need for MAP_WAIT/PIN_WAIT).
On ReactOS, by default, on mapping (and thus pinning), we will force a view
read so that data is in memory. The way our cache memory is managed at the
moment seems not to allow to fault on invalid access and if we don't force
read, the memory content will just be zeroed.
So trying to match Windows behavior, by default, now CcMapData() will enforce
the MAP_NO_READ flag and warn once about this behavior change.
CORE-12671
- The default format is used when no format name is specified: this is
the one we use so far in ReactOS:
<debug_class>:(<file>:<line>) <message>
with "debug_class" being "trace", "warn", "err".
- The "wine" format is the one used by Wine. It can be used when trying
to diff-compare traces for a module with the corresponding one
obtained from a Wine run. It can also be useful because the logging of
Wine-synced code assumes that the function names are automatically
added by the helper macros "FIXME()", "TRACE()", "WARN()" or "ERR()",
and not manually inside the logging string given to these macros:
for example:
FIXME("(%params) message\n", params);
displays:
fixme:<module>:SomeFunc(params) message
- The "extended" (or "ext") format is very noisy and tries to output a
lot of information; it is a hybrid of the previous two formats:
<debug_class>:(<file>:<line>):<channel>:SomeFunc <message>
Support for displaying the current process ID is added in
addition to the already existing support for thread ID.
Google Chrome with -no-sandbox parameter in ReactOS wouldn't display the web page because first-chance exception raised.
CORE-14926
- Remove FullName, Style, and FaceName members from TEXTOBJ structure.
- Add TextFace member into TEXTOBJ structure.
- Add MatchFontName() and MatchFontNames() helper functions.
- Fix GetTextFace() and related.
These macros should be implemented using the FsRtlTestAnsiCharacter() macro
that performs extended tests, especially for buffer overruns.
This fixes FsRtlIsAnsiCharacterWild(), FsRtlIsAnsiCharacterLegalFat(),
FsRtlIsAnsiCharacterLegalHpfs(), FsRtlIsAnsiCharacterLegalNtfs().
CORE-14067
Up to now, it was working by chance. Indeed, due to the invalid
ASCII check performed before calling FsRtlIsAnsiCharacterLegalHpfs(), the
macro is improperly called and overruns the FsRtlLegalAnsiCharacterArray
buffer. Fortunately, up to now, right after that buffer in kernel binary
there are strings which are more or less consistent with the flags that
are expected by the macro, causing a decent behavior of
FsRtlIsAnsiCharacterLegalHpfs() even for extended ASCII characters
(whereas FsRtlIsAnsiCharacterLegalHpfs() is only designed for ASCII
characters). But this is a totally out of control and wrong behavior.
A single change in the way the kernel was built could have caused the
CDFS driver not to work as previously.
I have made the choice to allow any extended ASCII character as done
for the unicode characters. This is a good compromise to avoid drastic
regressions for users having extended ASCII characters in their CD
file names.
This imports proposed upstream commit 1b6b625641dffb49951e60398e1a9c672318ea71
See pull request https://github.com/Microsoft/Windows-driver-samples/pull/278
CORE-14067
- Simplify the usage of the PUPDATE_PROGRESS callback.
- Add the possibility of specifying an initial non-zero StepCount when
creating the progress-bar (using the -Ex version), so that it can be
initially drawn with the expected initial count.
Of course ProgressSetStepCount() can continue to be used.
It's based on the code that was in CcPinRead() implementation. This
made no sense to have CcPinMappedData() doing nothing while implementing
everything in CcPinRead(). Indeed, drivers (starting with MS drivers)
can map data first and pin it afterwards with CcPinMappedData(). It was
leading to incorrect behavior with our previous noop implementation.
Added callback function that redirects calls to GreExtTextOutW that didn't went through lpk BiDi processing, calls that are from the kernel.
Completely solves JIRA issue CORE-6910.
And, don't emit \r\n when an \r alone is encountered.
This fixes the problem of extra newlines appearing when redirecting
"more" output to a file.
CORE-14592
Updated the font substitutions to use the new Hebrew glyphs in the updated Tahoma font.
Also added substitution for the Ubuntu font to make the Lautus and Modern themes legible with Hebrew UI elements.
Follow up to PR #789 and #790 ; CORE-14880, CORE-14943
Improve the progress bar functions:
- Add a member in the struct, which will be used as a bar colour parameter for the new CreateProgressBarEx().
- CreateProgressBar() which will now invoke the Ex variant without the additional parameter.
Short: The code was suffering from an off-by-one bug (inconsistency between inclusive end exclusive end address), which could lead to freeing one page above the initialization code. This led to freeing part of the kernel import section on x64. Now it is consistently using the aligned/exclusive end address.
Long:
* Initialization sections are freed both for the boot loaded images as well as for drivers that are loaded later. Obviously the second mechanism needs to be able to run at any time, so it is not initialization code itself. For some reason someone decided though, it would be a smart idea to implement the code twice, once for the boot loaded images, once for drivers and concluding that the former was initialization code itself and had to be freed.
* Since freeing the code that frees the initialization sections, while it is doing that is not possible, it uses a "smart trick", initially skipping that range, returning its start and end to the caller and have the caller free it afterwards.
* The code was using the end address in an inconsistent way, partly aligning it to the start of the following section, sometimes pointing to the last byte that should be freed. The function that freed each chunk was assuming the latter (i.e. that the end was included in the range) and thus freed the page that contained the end address. The end address for the range that was returned to the caller was aligned to the start of the next section, and the caller used it to free the range including the following page. On x64 this was the start of the import section of ntoskrnl. How that worked on x86 I don't even want to know.
This fixes a race condition, where the message thread was still running when InstallReactOS returned and syssetup.dll was unloaded by the caller (setup.exe).
Volume boot record is also implemented, it supports reading BTRFS tree structures with upto 64k node size.
This support required to change all path in Free Loader to lowercase for better performance.
CORE-13769
- Checks are actually uses-after-free, "by design" :-<
- Actual test code duplicates EngCreateSemaphore.c, EngAcquireSemaphore.c and EngReleaseSemaphore.c.
Normally CSRSS must not be connected to any winsta or desktop by default. It should manually connect
to a winsta/desktop only when it has to do some GUI operations, and then disconnect afterwards.
[USERSRV] Temporarily hackfix the harderror dialog display to the current input desktop.
as documented in "Advanced Windows NT" by Jeffrey M. Richter (Microsoft Press),
and in https://is.muni.cz/el/1433/jaro2010/PB167/um/cv5/undocumented_CreateProcess.pdf .
[INCLUDE][SERVICES][WIN32K:NTUSER] Add an undocumented STARTF_INHERITDESKTOP flag
for the STARTUPINFO::dwFlags structure member, whose purpose is to tell Win32k
that the created handles to the window station and desktop to which the
process is connecting to, can be inherited by its child processes.
It is used when starting interactive services.
Observed via API monitoring on Windows 2003.
CORE-11933 and PR #621.
Since this API is also called from WINSRV when calling the AllocConsole() API,
it can be tested more-or-less easily. The internal helper IntResolveDesktop()
is also tested during process connection to a window station, when such process
first calls a USER32 or GDI32 function.
This is also the functionality tested by the user32:desktop apitest.
- Adjust how IntResolveDesktop() is called.
CORE-11933 and PR #621.
We are potentially going to modify the window station name to be
created or opened, by one with the format "Service-0x<luidhigh>-<luidlow>$",
in case the user provides an empty name.
Since we want to use the user-mode ObjectAttributes structure pointer so
that the Object manager can correctly perform the access checks and the
capture, we actually need to construct the new window station name in the
user memory space! This allows us then to fetch the new name in the
ObjectAttributes structure so that Ob can use it.
All of this is performed under SEH.
For NtUserOpenWindowStation(), we also need to be Windows-compatible and
detect whether user-mode gave instead the special "Service-0x00000000-00000000$"
name (observed via API monitoring) that is used when one tries to open a
window station with empty name.
CORE-11933 and PR #621.
- Remove the related hack-FIXMEs;
- Adjust NtUserGetObjectInformation() in accordance.
- Retrieve the window-station/desktop object type string in NtUserGetObjectInformation()
also from the NT Object's header.
Also simplify the UOI_FLAGS case of NtUserGetObjectInformation() by reading
the handle inheritance information directly from the OBJECT_HANDLE_INFORMATION
structure returned by ObReferenceObjectByHandle().
Observed via API monitoring.
This corresponds to the case where Win32k creates a window station whose name is based on the logon session identifier for the calling process.
Add also a note about the fact that we need to use a per-session-based WindowStation directory name, as done already in Win32k.
CORE-11933 and PR #621.
* Remove unused cmake modules - Both are unedited versions of modules provided by cmake itself
* Remove Compiler/GNU.cmake - Only chang was various _INIT flags, which are handled
via CMAKE_USER_MAKE_RULES_OVERRIDE instead
* Remove Platform/Windows.cmake - There's no clear explination for this file being in reactos
and is simply an old version of the one in cmake
* Remove Platform/Windows-MSVC.cmake - _INIT variable changes moved to overrides-msvc.cmake
Remove /implib from link commands
* Remove CMakeDetermineASMCompiler.cmake - Only change from 3.2 is the addition of a compiler list for the generic ASM dialect, but toolchain files explicitly set a compiler so the list is never used
When we unselect text after the WM_CBLOSTTEXTFOCUS message,
make sure we also forget we have been focused at all;
otherwise the edit may become focused again, but with an
empty text selection.
CORE-10266
Spinning here is supposed to improve performance by avoiding scheduling
overhead, but because ReactOS's Win32 subsystem does not deal with this well,
it causes the system to freeze on ROS.
As a workaround, we disable spinning and jump directly to the wait.
It should become obvious when this starts actually causing performance issues
again and it's time to undo.
When we're about to close a file (ie, forget everything about it
and release any associated structure), actually delay it.
This allows keep data fresh in memory for faster reuse in case
it would be required. The effective closing will only happen after some time.
For specific operations, this will produce a real speed up in ReactOS.
For instance, with that patch, Winamp starts within seconds, instead of dozen
of minutes.
In most cases, it will bring ReactOS to performances it had before fixing
the huge leak in FastFAT (commit 94ead99) without leaking the whole FS.
For now, due to regressions, this is only activated for files and not
for directories. Once it gets fixed, it will be enabled for both.
CORE-14826
CORE-14917
This happened for the livecd, where the cdfs driver was inserted twice - once since it is the driver for the boot device, and the 2nd time, because it was specified in the registry - which was then initialized twice by the kernel, leading to a name conflict when trying to create another device object, which resulted in the newly created device object to be dereferenced again and the driver unloaded. This can be seen from the debug message "(ntoskrnl\mm\ARM3\sysldr.c:955) Leaking driver: cdfs.sys"
On x64 we only map 1GB of pages, so adjust MM_MAX_PAGE accordingly and also respect that value when searching for the best location of the page lookup table.
CORE-11048 #resolve
The PROCESS_DEVICEMAP_INFORMATION union has 2 fields, one is a handle, the other one is a structure of 36 bytes (independent of architecture). The handle forces 64 bit alignment on 64 bit builds, making the structure 4 bytes bigger than on 32 bit builds. The site is checked in NtQueryInformationProcess (case ProcessDeviceMap). The expected size on x64 is the size of the Query structure without alignment. autocheck correctly passes the site of the Query union member, while smss passes the full size of PROCESS_DEVICEMAP_INFORMATION. Packing the structure is not an option, since it is defined in public headers without packing. Using the original headers sizeof(PROCESS_DEVICEMAP_INFORMATION) is 0x28, sizeof(PROCESS_DEVICEMAP_INFORMATION::Query) is 0x24.
- Rename ObDirectoryType to ObpDirectoryObjectType and remove it from NDK (this is not exported!)
- Rename ObSymbolicLinkType to ObpSymbolicLinkObjectType
- Remove duplicated ObpTypeObjectType from ob.h
fix control overlapping in IDD_FOLDER_OPTIONS_FILETYPES
(broke in 0.4.10-dev-24-ed1c6bb0060d06f4c0167d9c2d732f4dc2b4cce4)
and translate IDD_NEWEXTENSION, IDD_EDITTYPE, IDD_ACTION, IDD_FOLDER_CUSTOMIZE
Additional changes:
- Adapt the message-box dialog procedure to correctly take into account
the presence or absence of a CANCEL button and take appropriate actions.
- Cache the strings retrieved by MB_GetString() (add FIXME notices too).
- Remove usage of ButtonLen and ibtn-pointer arrays, to make the code
more generic and less dependent on the (maximal) number of buttons
in the message boxes.
- Move around the code that computes the positions and sizes of the
window controls.
Kernel stacks that re freed, can be placed on an SLIST for quick reuse. The old code was using a member of the PFN of the last stack page as the SLIST_ENTRY. This relies on the following (non-portable) assumptions:
- A stack always has a PTE associated with it.
- This PTE has a PFN associated with it.
- The PFN has an empty field that can be re-used as an SLIST_ENTRY.
- The PFN has another field that points back to the PTE, which then can be used to get the stack base.
Specifically: On x64 the PFN field is not 16 bytes aligned, so it cannot be used as an SLIST_ENTRY. (In a "usermode kernel" the other assumptions are also invalid).
The new code does what Windows does (and which seems absolutely obvious to do): Place the SLIST_ENTRY directly on the stack.
This commit adds support for the MS symbol encoding (FT_ENCODING_MS_SYMBOL) in our font engine.
It also fixes the display of the Anastasia font. And replacing our Marlett font with Windows' Marlett font is also fixed.
CORE-13269, CORE-14907
For whatever (unknown) reason yet, this stream file object
seems to be deleted when still in use while installing
ReactOS on BtrFS partition, leading to use after free.
So, quick and dirty hack: leak it to prevent deletion.
CORE-13769
Fix vertical alignment and Aspect penalty; improve font penalty system; fixes for Wine tests.
This is an update of #706. See CORE-11536 for more details.
The display of the following programs is fixed:
- Total Commander 8.52 setup: font displayed too large - CORE-11620.
- Effective File Search 6.8.1 german localization text rendering issues - CORE-14378.
- Font garbage in register splash screen in Foxit Reader 7.1.5 - CORE-9767.
- Calipers-1 is not displayed correctly - CORE-14302.
- Some MSI-installers draw their dialogs too large (example: Click-N-Type Virtual Keyboard 3.03.0412) - CORE-13161.
- Irfanview 4.50: font in zoom combobox displayed too large - CORE-14396.
- Rufus: The window and controls are displayed larger than necessary - CORE-14461.
It's hardly understandable and doesn't really makes sense.
Furthermore, it breaks compatibility with 3rd party FSD that
don't implement such FSCTL.
Obviously, Windows doesn't do this.
Particular DIR commands like: "DIR .", "DIR .." now work as expected,
and we also correctly fix the behavior for files without extension,
that r38746 (2b06cfc0) originally tried to fix but broke the previous
examples.
Therefore "DIR *." and "DIR noextfile." work too.
Pathological cases like "DIR \...", "DIR \...\.", "DIR ..\...\.." and
the like (and with more than 3 dots) now work as expected.
Adapted from PR #592 by Katayama Hirofumi MZ, but with extended bugfixing.
CORE-13961
- Service bits are set by NetrServerSetServiceBits and can be retrieved by NetrServerGetInfo.
- The real name of function 42 is NetrServerSetServiceBitsEx and the real name of function 47 is NetrDfsSetServerInfo.
The "Run" dialog failed when multiple parameters wee specified.
We use instead a newly-created ShellExecCmdLine() helper function to fix the problem (as found via API-tracing on Windows).
Note that ShellExecCmdLine() starts to be exported with Vista+.
- Implement ShellExecCmdLine() function in shell32.
- Add URL support.
- Fix RunDlgProc function in shell32.
- Add a testcase for ShellExecCmdLine() function.
CORE-14790
- Implement EnumSystemFirmwareTables
- Implement GetSystemFirmwareTable
These functions currently using registry workaround and can be improved later.
CORE-12105
I found it unintuitive that each time I selected some software for installation, it wasn't automatically deselected after installation. I kept re-triggering the same installation a few times before I realized I had to manually uncheck the checked packages.
- Add support for the following command-line arguments: /k, /p (and a little bit
of /r), and add support for checking a particular volume.
A description of these arguments can be found at:
"Description of Enhanced Chkdsk, Autochk, and Chkntfs Tools in Windows 2000",
https://web.archive.org/web/20150215210228/http://support.microsoft.com:80/kb/218461
- Slightly improve some messages.
- Get rid of legacy OpenDirectory() and simplify GetFileSystem().
For user mode, when probing output buffer, if it's null, length
will also be set to 0.
This avoids user mode applications being able to trigger various
asserts in ReactOS (and thus BSOD when no debugger is plugged ;-)).
Also, improve NtUserSetWindowStationUser() capture order, make psid optional as it should (and avoid a user-mode triggered BSOD), and initialize luidUser only when everything succeeded.
GAS uses ".double" symbol for declaring floating-point constants and
".quad" symbol for declaring 64-bit numbers.
This is not compatible with our macro for MASM and introduces bugs.
Now 64-bit constants are supposed to be declared using ".quad" macro.
NDK: Define PLUGPLAY_CONTROL_PROPERTY_DATA.Properties and PLUGPLAY_CONTROL_DEVICE_RELATIONS_DATA.Relations values.
NTOSKRNL: Map PLUGPLAY_CONTROL_PROPERTY_DATA.Properties values to IoGetDeviceProperty properties and add (dummy) code for unsupported cases.
UMPNPMGR: Use PLUGPLAY_CONTROL_PROPERTY_DATA.Properties values in PNP_GetDeviceRegProp.
An owned popup window should be hidden when its owner window was minimized.
- Add IntWinListOwnedPopups function.
- Fix ShowWindow and ShowOwnedPopups functions.
CORE-14818
See also: CORE-3326, CORE-12252, CORE-13168, and CORE-14824.
The keyboard shortcuts Win+D and Win+M are also enabled.
- Implement IShellDispatch4::ToggleDesktop().
- Implement some commands in CTrayWindow.
- Add "sdk/include/reactos/traycmd.h" for tray commands.
- Fix task window switching.
- Improve the user32!SwitchToThisWindow() function and use it.
CORE-14318, CORE-13157
See also: CORE-14806 and CORE-8723
- Implement shell32 RegenerateUserEnvironment() function and use it.
- CShellBrowser and CDesktopBrowser implements WM_SETTINGCHANGE actions.
CORE-1459, CORE-14397
This is required, since newer versions of MSVC demand that non-member operator new/delete are in the global namespace and neither static nor inline. See https://msdn.microsoft.com/en-us/library/mt723604.aspx ("Overloaded operator new and operator delete")
- Plus: Add missing email address in 74f92c3d86.
- Remove superfluous blank lines.
- Add 2 TRACE("FunctionName()\n").
- Use "0x%llx", not "%lu", for BaseAddress and Length. (Copypasta in my recent commits.)
- Add "0x" to 1 "%p" and 1 "%lx".
This was a regression from last Wine (Staging) 3.9 sync (f68b0c51) that
introduced the usage of this new dialog.
To do: Properly generate the dialog localization.
CORE-14656, CORE-14674, CORE-14689, CORE-14769
* Latest VS does not like half translated files :/
* errcodes.mc is duplicated in kernel32
* It would probably be better to separate each language into it's own file, but we need a build mechanism for that
- Fix a regression introduced in r65415 (1795a3bf) where the directory
paths stored in "Windows\CurrentVersion\Explorer\User Shell Folders"
were stored in expanded format instead of in unexpanded format:
_SHGetDefaultValue() *MUST* return unexpanded paths by design!!
- Augment _SHExpandEnvironmentStrings() and _SHGetUserShellFolderPath()
to take a user token handle to be able to correctly resolve/expand
user-specific directory paths.
- Fix _SHExpandEnvironmentStrings() so that it always retrieve the
correct current user / all-users directory paths by calling userenv
functions, instead of hardcoding a broken logic. As a result this
removes the "C:\Documents and Settings\SYSTEM" ghost directory we got
during ReactOS installation.
- Delimit the changes with respect to Wine by #if(n)def __REACTOS__ .
This effectively removes the temporary hacks introduced in r73486 (a35a785b)
and in r73487 (0ce031f7) for the LocalService and NetworkService accounts.
Now these accounts (and their corresponding user profiles) are created
on the fly, the first time a service that needs these is started.
The code introduced in r73501 (ade0d2cd) is now re-enabled: this has
been made possible thanks to commit f42b4bbe (thanks Eric!).
Dedicated to Joachim Henze ;-)
CORE-12541, CORE-12279
Refactor the font engine code in "win32ss/gdi/ntgdi/freetype.c".
- Add "g_" prefix to global variables.
- Make the locking macros function-like.
CORE-13065
- Overhaul SepCreateToken() and SepDuplicateToken() so that they
implement the "variable information area" of the token, where
immutable lists of user & groups and privileges reside, and the
"dynamic information area" (allocated separately in paged pool),
where mutable data such as the token's default DACL is stored.
Perform the necessary adaptations in SepDeleteToken() and in
NtSetInformationToken().
- Actually dereference the token's logon session, when needed, in the
'TokenSessionReference' case in NtSetInformationToken().
- Overhaul SepFindPrimaryGroupAndDefaultOwner() so that it returns
the indices of candidate primary group and default owner within the
token's user & groups array. This allows for fixing the 'TokenOwner'
and 'TokenPrimaryGroup' cases of NtSetInformationToken(), since the
owner or primary group being set *MUST* already exist in the token's
user & groups array (as a by-product, memory corruptions that existed
before due to the broken way of setting these properties disappear too).
- Lock tokens every time operations are performed on them (NOTE: we
still use a global token lock!).
- Touch the ModifiedId LUID member of tokens everytime a write operation
(property change, etc...) is made on them.
- Fix some group attributes in the SYSTEM process token, SepCreateSystemProcessToken().
- Make the SeCreateTokenPrivilege mandatory when calling NtCreateToken().
- Update the token pool tags.
- Explicitly use the Ex*ResourceLite() versions of the locking functions
in the token locking macros.
- Use TRUE/FALSE instead of 1/0 for booleans.
- Use NULL instead of 0 for null pointers.
- Print 0x prefix for hex values in DPRINTs.
- Use new annotations for SepCreateToken() and SepDuplicateToken().
Caught while debugging, in the case the ImpersonationLevel value was
uninitialized, due to the fact it was left untouched on purpose by
PsReferenceEffectiveToken().
- Use a common helper function used by both the ANSI and UNICODE functions
so that the large code block that deals with setting the new process token
becomes automatically common to both these functions, while the ANSI vs.
UNICODE conversions are handled directly via the CreateProcessA/W() calls.
- Duplicate the token and acquire the process primary token assignment
privilege before calling NtSetInformationProcess().
Function TRASH_CanTrashFile() would always fail because GetVolumeInformationW() requires only the base root path. The path (stored in buffer wszRootPathName) was not being stripped correctly.
CORE-12340
- Improved Italian translation for USETUP, WELCOME, REACTOS, EXPLORER.
- Completed translation in SYSSETUP.
- Correct some characters encoding and text alignment in USETUP.
"warning: cast between incompatible calling conventions 'cdecl' and 'fastcall'; calls through this pointer may abort at runtime [-Wcast-calling-convention]"
CORE-14306
- "(sdk/lib/rtl/heappage.c:1329) corrupted suffix pattern"
detected at
"dll/win32/syssetup/wizard.c:2672 (InstallWizard)"
- Add an ASSERT(), as hinted by Mark Jansen.
Import the following Wine commits:
* 79636bebbfa Michael Müller: advapi32: Set last error to ERROR_SUCCESS in GetSidIdentifierAuthority.
* 1d03ba76116 Hans Leidekker: advapi32: Fix parsing empty DACL/SACL security descriptor strings.
* 5bc2e83c7ab Hans Leidekker: advapi32: Fix size returned from ConvertStringSecurityDescriptorToSecurityDescriptor on 64-bit.
"reactos-cov\modules\rostests\winetests\msi\msi_winetest.rc(2): error RC2135: file not found: custom.dll [C:\ros_build\modules\rostests\winetests\msi\msi_winetest.vcxproj]".
Output files are written to Debug\ and Release\ subfolders in VSSolution mode, so the RC compiler will only find custom.dll if it knows to look in the right folder.
Thanks to Thomas Faber, who helped.
CORE-11836
Otherwise fall back to the computed argv[0].
This is expected by some applications, for example Git.
Code is adapted from Wine.
Many thanks to Stanislav Motylkov for having investigated this bug!
CORE-12931 CORE-13892 CORE-13898 CORE-14066
Multiple "error C4189: '[...]': local variable is initialized but not referenced".
Thanks to Yuntian Zhang, who hinted at this in CORE-11836.
CORE-11280
Implement mshta.exe. It links directly to the function RunHTMLApplication in mshtml and passes the file name and other optional arguments along. While the function is unimplemented in mshtml, it's a start.
[MSHTML] Add ADD_IMPORTLIB to CMakeLists so mshta can link to mshtml.
[BOOTDATA] Add the association information for mshta.exe to the registry.
kmtest:NtCreateSection calls CcInitializeCacheMap with a
NULL value for SectionObjectPointers. This will cause an exception when
trying to access it, which in Windows can be handled gracefully.
However accessing it while holding ViewLock means the lock will not be
released, leading to an APC_INDEX_MISMATCH bugcheck.
This solves the problem by allocating SharedCacheMap outside the lock,
then freeing it again under lock if another thread has updated SharedCacheMap
in the mean time. This is also What Windows Does(TM).
DIRECTORY_ALL_ACCESS is not the correct flag, because it is used with the NtCreateDirectoryObject API
that creates virtual "directories" in the NT object namespace; that's not our purpose here.
svn path=/branches/setup_improvements/; revision=74716
- Use NT string safe functions.
- Add support for other bootloaders;
- Update a couple of comments;
- Remove deprecated __REACTOS__ #ifdefs.
svn path=/branches/setup_improvements/; revision=74713
- Call OpenAndMapFile() with its new ReadWrite parameter;
- Add informative comments in osdetect.c;
- In partlist.c, check whether a disk has a valid MBR by also checking for its 0xAA55 signature.
svn path=/branches/setup_improvements/; revision=74712
- In DoesFileExist(): Call NtOpenFile with FILE_GENERIC_READ instead of the more generic GENERIC_READ access right.
- OpenAndMapFile(): Add support for opening & mapping files with write access (to be used latter).
svn path=/branches/setup_improvements/; revision=74710
- Split IniCacheLoad() and IniCacheSave() into: themselves & IniCacheLoadByHandle() and IniCacheSaveByHandle(),
respectively, so that we can load & save INI files if we already have an opened handle to them.
svn path=/branches/setup_improvements/; revision=74711
- Remove a redundant NtClose() call;
- Return failure if NtQuerySymbolicLinkObject() fails;
- Use RTL_CONSTANT_STRING and RtlInitEmptyUnicodeString() where needed;
- Reduce code indent level;
- Add old-style function annotations;
- Remove the deprecated code copyright notice, since the copyright in usage is already reported in the COPYING file in the top level ReactOS source code directory.
svn path=/branches/setup_improvements/; revision=74698
So far we only have:
- a function FindNTOSBootLoader() that detects the existence of a given boot loader;
- a function EnumerateNTOSBootEntries() (and corresponding helpers) that enumerate the different boot entries in the configuration file(s) for a given boot loader, and for each entry, calls a user-provided callback.
Only supported at the moment: ntldr and freeldr.
Doing that allows me to simplify large portions of the NT-OS detection code so that it becomes more bootloader-agnostic, and this will help me for simplifying some parts of usetup/bootsup.c too, later...
svn path=/branches/setup_improvements/; revision=74661
- Adjust also its callers, adjust OpenAndMapFile() parameters.
- Related to that, simplify IsValidNTOSInstallation() parameters & introduce a IsValidNTOSInstallation_UStr()
that does the same, but takes a UNICODE_STRING instead.
- Simplify CheckForValidPEAndVendor().
Now only exactly 5 calls use the "old" 'DoesFileExist' syntax, using a temporarily auxiliary function "DoesFileExist_2"...
svn path=/branches/setup_improvements/; revision=74641
This allows building concatenated paths with an arbitrary number of separated components.
- Use the newly-introduced CombinePaths() and ConcatPaths() functions.
- Fix also few comments, and place some UNICODE_NULLs here & there.
svn path=/branches/setup_improvements/; revision=74637
svn path=/branches/setup_improvements/; revision=74640
- filesup.c's functions ConcatPaths(), Does[Path|File]Exist(), NtPathToDiskPartComponents(), OpenAndMapFile(), UnMapFile();
- Move the inicache library to setuplib as it'll be used for the 1st stage GUI setup too (indeed, there is no good INI file API
under Win32; the Win32 profile "API" is just good enough to manipulate the win16 ini files, and are here anyways for backward
compatibility purposes only);
- Move the OS detector too.
- Remove the duplicated ConcatPaths() code in arcname.c.
svn path=/branches/setup_improvements/; revision=74634
svn path=/branches/setup_improvements/; revision=74638
Added to Development Team:
Aleksandar Andrejevic
Alexander Shaposhnikov
Amine Khaldi
Benedikt Freisen
Cameron Gutman
David Quintana
Giannis Adamopoulos
Hermès Bélusca-Maïto
Jérôme Gardou
Katayama Hirofumi MZ
Mark Jansen
Peter Hater
Stanislav Motylkov
Thomas Faber
Vadim Galyant
Added to Release Engineers:
Amine Khaldi
Joachim Henze
Thomas Faber
Addendum to 16d6c9b5e0
Giannis suggested and Thomas explained:
Before the Wine sync, comctl32/button.c was based on ROS's
user32/button.c, so _USER32_ would highlight changes between
user32 (originally based on Wine) and comctl32. Now our
comctl32/button.c is based on Wine's comctl32/button.c, so
__REACTOS__ is enough.
- Introduce two small helpers to change and restore the console title.
- Console title can change even when internal commands are executed.
- Note that when commands are run from within batch files, title is unchanged.
- When "cmd.exe /c command" is run, the console title is unchanged; however
when "cmd.exe /k command" is run, the console title changes.
- Store the full profile user name in the listview item for later use in the delete and copy functions.
- Change the button state according to the selected profile.
- Fix some hard-coded string sizes.
Our implementation had a bug that could be triggered while
building our USBD library on ReactOS: the compare function
could be called with a NULL pointer instead of a valid value.
With this bug fixed (and the cmd hack in CORE-14648), ReactOS
can totally selfhost :-)! I was able to build LiveCD and BootCD
without any trouble, crash, deadlock or whatever.
(Next step: having a buildbot slave hosted on ReactOS ;-)).
Enjoy: https://twitter.com/HeisSpiter/status/1002880397103988737
CORE-14680
Fixes a regression that was introduced by Wine Staging 3.3 Sync.
Symptom was missing button text and icons for specific buttons
in Delphi applications, e.g. Double-Commander, Lazarus, Peazip.
The patch restores and ifdef-guards previous changes of Giannis.
- In the partlist.c disk getters: Remove useless "IsListEmpty(&List->DiskListHead)" checks, because this is actually the kind of check the while() loop does just after...
- Fix few DPRINTs.
svn path=/branches/setup_improvements/; revision=74629
- Improve FindExistingNTOSInstall() so that we can find an existing installation either by system root
ARC path or NT path. This is used during the enumeration of available installations from the boot.ini/freeldr.ini
and during other existence & validity checks of NTOS installations.
- Improve AddNTOSInstallation() so that we can save the system root ARC path and NT path of the installation,
as well as its partition entry structure pointer, for caching & later retrieval purposes.
- Remove some deprecated comments & todos, and implement other todos.
- Improve the output of some DPRINTs.
- Fix the return value of FindSubStrI.
svn path=/branches/setup_improvements/; revision=74632
- isspace('\0') returns FALSE anyways so no need to separately test for a NULL character;
- The (str/wcs)toul function cannot return a NULL pointer from its second paramter;
- VersionInfo32_FindChild(): the third argument is indeed a number of characters (not bytes),
so rename the parameter to make this fact clear. The function is however correctly used within this module.
svn path=/branches/setup_improvements/; revision=74629
In all the disks/partitions available, it searches for the presence of freeldr.ini / boot.ini, open & parse them,
and enumerates the available boot entries (as candidates for installations). For each of them, it maps their ARC paths
into the NT namespace (hence the ARC 2 NT path resolver committed in r74621), then attempts to detect in these paths
the existence of NTOS installations.
svn path=/branches/setup_improvements/; revision=74622
The NT path resolver allows mapping between an ARC path as specified in freeldr.ini / boot.ini , to its corresponding NT path, if possible.
Currently, only the mapping direction "ARC to NT" is implemented. It will be used wherever such mappings are needed, for example when identifying
the ReactOS / Windows installations from the available freeldr.ini / boot.ini entries (for upgrading / repair purposes).
The resolver supports the usual ARC paths: multi()disk()[r|f]disk()[partition()] ; eisa()disk()[r|f]disk()[partition()] ; multi()disk()cdrom() ;
scsi()disk()[r|f]disk()[partition()] ; scsi()cdrom()fdisk() ; ramdisk(x) ; net(x) (actually reported as "unsupported" since it would map to some
path on some network), and the newly-introduced Win2k signature()disk()rdisk()[partition()].
The code is in work-in-progress status.
Some validation tests, that were used during the implementation of the resolver, have been added.
svn path=/branches/setup_improvements/; revision=74621
svn path=/branches/setup_improvements/; revision=74631
The aim is to use this detector to be able to detect and select an existing installation of ReactOS for upgrading.
The user then could either select one, or skip this step and perform a regular ReactOS installation.
What remains to be done, is to parse the NTOS loader configuration files (freeldr.ini in ReactOS' case, or boot.ini in Win2k3's case, etc...)
to retrieve the actual installation paths. So far these are currently hardcoded for testing purposes only.
The detector attempts to distinguish between ReactOS and Windows installations by checking at the company name vendor of the ntoskrnl.exe & ntdll.dll files,
so that only ReactOS installations are allowed to be upgraded.
svn path=/branches/setup_improvements/; revision=74527
svn path=/branches/setup_improvements/; revision=74550
In working on some problems with Time Zone adjustments in ReactOS I found that the RtlQueryTimeZoneInformation function is partially broken and wrote an apitest to capture the failures.
Patch by Doug Lyons.
CORE-14658
This allows the user to edit the File Type / extension information by clicking 'Advanced' button in the 'File Types' property sheet of Folder Options.
CORE-12906, CORE-10972
Introduce BiDi (bi-directional text) support for ExtTextOut and GetCharacterPlacement, using Wine's GDI BIDI_Reorder function.
Solves the main issue with CORE-7003.
To be compatible with Win2k3+, introduce the "Language Pack" (LPK) dll.
- All the bidi code is removed from gdi32 and replaced by calls to LPK.
Gdi32 uses dynamic linking to lpk.dll. In case of linking failure no bidi processing will be available.
- Implemented LpkGetCharacterPlacement.
- Implement LpkExtTextOut.
- Add a demo test program to show how the apis should function.
- Added all the remaining code, added special case for lpDx calculation if also GCP_GLYPHSHAPE flag was called.
Applications that call GCP that use GCP_GLYPHSHAPE flags should also use the GCP_REORDER flag.
(As written in https://msdn.microsoft.com/en-us/library/windows/desktop/dd144860(v=vs.85).aspx )
- Add ETO_RTLREADING flag handling.
Imported the ETO_RTLREADING flag handling from wine, which changes the string part order (runs).
A RRR1LLLRRR2 string without will show as RRR1LLLRRR2 without it, with it RRR2LLLRRR1.
From the existing IniCacheLoad() function, introduce a IniCacheLoadFromMemory() function that just does the same (initialize an INI file cache and parse the INI file), but takes the input from a memory buffer. Then, rewrite the IniCacheLoad() function to just open the file given in input, and then fall back to calling IniCacheLoadFromMemory.
The IniCacheLoadFromMemory() function will be used later.
svn path=/branches/setup_improvements/; revision=74620
- Add a NtPathToDiskPartComponents() helper, that takes in input a fully qualified NT path to a file on hard disk,
e.g.: \Device\Harddisk1\Partition2\foo\bar, and returns in output the disk number ('1'), the partition number ('2'),
and the the path component "\foo\bar" that is after the device-harddisk-partition identifier.
- Make the OpenAndMapFile() return the file size of the opened file.
Both of these additions will be used soon.
- Turn a isspace() call into a iswspace() one.
svn path=/branches/setup_improvements/; revision=74619
- Use the previously-introduced 'PreparePartitionForFormatting()' function; set the FormatState of the newly-formatted partition to Formatted.
- Remove the hackish call to CreateFileSystemList() in SelectFileSystemPage().
- Move 'TempPartition' and 'FormatState' back to USETUP.
- Put large "case"-blocks into brackets.
svn path=/branches/setup_improvements/; revision=74575
svn path=/branches/setup_improvements/; revision=74576
- Add a PreparePartitionForFormatting routine that sets the partition ID depending on the chosen filesystem.
- The 'FORMATMACHINESTATE FormatState' machine-state and the 'TempPartition' members of the partition list structure is purely a USETUP convenience, so remove them from the PARTLIST structure and move them back into USETUP.
- Attempt to recognize the filesystem (set the 'FileSystem' member of PARTENTRY) of partitions we are adding into the PARTLIST list.
- Fix the return value of the SelectPartition function, which is by the way completely broken (it doesn't do what it is supposed to do; alternatively its naming is completely wrong...).
svn path=/branches/setup_improvements/; revision=74572
svn path=/branches/setup_improvements/; revision=74573
- Create the beginnings of a "setuplib" library, whose aim is to be shared between the (currently existing) 1st-stage text-mode installer, and the (future) 1st-stage GUI installer.
- Finish to split the GenList and PartList codes into their UI part, which remain in usetup, and their algorithmic part, which go into setuplib.
- Move SetMountedDeviceValue into the PartList module.
- Split the FileSystem list code into its UI and the algorithmic part (which goes into setuplib under the name fsutil.c).
* The algo part is meant to be able to manage the filesystems available on the running system, similarly to what is mostly done (in scattered form) in fmifs, format, chkdsk / autochk codes...
It also manages the partition filesystem recognition, using OS routines.
* The UI part manages the FS list as it appears on screen, showing only the possible FSes that can be used to format the selected partition (a bit similar to what we do in the shell32's drive.c, etc...).
- Adapt the calling code to these changes.
- Remove some "host" code that was dating back from the dark old times.
svn path=/branches/setup_improvements/; revision=74570
svn path=/branches/setup_improvements/; revision=74659
This will allow to reuse it for the 1st-stage GUI setup too, while using another UI representation.
Add also two partition iterator functions: GetNextPartition and GetPrevPartition.
svn path=/branches/setup_improvements/; revision=74554
The aim here is to decouple the UI-specific code from code that can be used by both the text-mode USETUP and a future 1st-stage GUI setup.
Indeed, the GenLists can actually be used in the 1st-stage GUI; and their contents be displayed inside ListBoxes/ListViews... (this is just one example amongst others).
Additionally (in usetup.c):
- Make both FormatPartitionPage and CheckFileSystemPage return PAGE_NUMBERs.
- Improve a couple of comments.
svn path=/branches/setup_improvements/; revision=74553
- Allow retrieving the number of items existing in the list;
- Fix the name of some function parameters.
svn path=/branches/setup_improvements/; revision=74547
- Document more some of the fields in the PARTENTRY, DISKETNRY and PARTLIST structures;
- Remove the redundant members "SystemDisk", "OriginalSystemDisk" and "TempDisk" in PARTLIST as these can be consistently deduced from the corresponding (Original)(System)(Temp)Partition members
(note that we however keep "CurrentDisk" alongside "CurrentPartition", see the comment in the code why we do it so).
- Adjust the rest of the code to take the removal of the redundant members into account. The 2nd parameter of GetNextUnformattedPartition() and GetNextUncheckedPartition() is now really optional.
- Introduce a SetPartitionType() helper to simplify the code that sets the partition type, which also automatically adjusts other internal variables of said partition in accordance.
- "Mounted" logical drives can have assigned letters too, registered in \DosDevices\.
svn path=/branches/setup_improvements/; revision=74532
- Use explicit ansi string safe functions (where they are used ansi, but not explicitely);
- Add (old-school) function parameters annotations;
- Use PARTITION_ENTRY_UNUSED where needed (instead of hardcoding its value);
- Turn some functions static to this module;
- Turn the 2nd parameter of both GetNextUnformattedPartition() and GetNextUncheckedPartition() optional (for next commit);
- Improve some comments;
- Use NT types.
svn path=/branches/setup_improvements/; revision=74531
- Introduce code that detects the filesystem of a (mounted?) partition, using NtQueryVolumeInformationFile() with FileFsAttributeInformation class,
aka. rely on ReactOS itself (kernel, storage stack, filesystem drivers...) to recognize the FS on a partition that should normally be seen by the system.
This currently half-works for whatever reason (to be insvestigated), while it works on Windows.
- Fix few comments & a function parameter name.
- Use NT string pointer types.
svn path=/branches/setup_improvements/; revision=74529
- ConcatPaths that concatenates paths (or a path and a file name);
- OpenAndMapFile (resp. UnMapFile), whose purpose is to open a file and map it in memory (resp. unmap it from memory).
- Add extra optional parameters to DoesPathExist and DoesFileExist: an optional "RootDirectory" handle and, for DoesFileExist only, an optional PathName.
- Close the opened file handles only on success.
svn path=/branches/setup_improvements/; revision=74528
svn path=/branches/setup_improvements/; revision=74538
svn path=/branches/setup_improvements/; revision=74549
- Implement AcquireSpinlock, ReleaseSpinlock and GetExtendedFunctionTable notifications.
- Implement a bus scan routine, borrowed from scsiport.
Storport and storahci are now able to detect a disk device attached to a Virtual Box AHCI controller.
This avoids a really nasty race condition in our cache controler where
two concurrents could try to initialize cache on the same file.
This had two nasty effects: first shared map was purely leaked and erased
by the second one. And the private cache map, allocated on the first shared
cache map couldn't be freed and was leading to Mm BSOD (free in a middle of
a block).
This was often triggered while building ReactOS on ReactOS (with multi threads).
With that patch, I cannot crash anylonger while building ReactOS.
CORE-14634
- Get rid of DoAutoLogon() and the AutoLogonState variable.
- Handle the auto logon in the normal logon code path instead.
- Set bAutoAdminLogon to FALSE on a successful login. This will prevent another automatic logon after a logoff.
- The IgnoreShiftOverride feature got lost by this change but I will add it again.
This avoids race conditions under high IOs and thus corruption on
the FS, or assertions failures in the kernel.
Easily triggered by building ReactOS on ReactOS ;-).
Until now, our support for dirty volumes was totally broken
to a point where, on FAT32 volume, the dirty couldn't even
be written nor read from the disk.
This commit totally rewrites its handling, for both FAT16 and FAT32
so that it's now fully functionnal. Furthermore, it also gets
totally compatible with our vfatlib, and thus, autochk.
Now, on mount, FastFAT will check if the volume is dirty or not, and
autochk will be able to ask for a repair if dirty. vfatlib will
repair the volume and remove the dirty bit. So that, on next
reboot, the volume will be mounted clean.
As a reminder, the dirty bit is set immediately after mounting
the volume, so that, if you crash or have a powercut, autochk
will always attempt to repair your volume (with more or less,
that's FAT!).
If you want to experience without breaking your FAT volume,
just boot, open a cmd prompt and type: fsutil dirty set c:
and reboot!
CORE-13758
CORE-13760
CORE-13759
This is a hack, and totally not the default behavior.
But it will allow autochk locking the boot volume in
order to verify it on boot, in case it would have
been improperly dismounted.
CORE-13759
This avoids initializing cache directly on directory
open/create.
The advantage is we reduce the load on cache manager
and on memory manager by avoiding creating everytime
a stream file object, and initializing cache for it.
This will avoid initializing cache for started
applications 'current directory' which is just opened
for having a valid handle but no read/write is performed
in it, by default.
This is a step forward for autochk.
CORE-14629
This is cleaner behavior and lets report failures.
- Create KmtEnsureServiceState(),
called at end of KmtStartService() and KmtStopService().
- Create KmtGetServiceStateAsString(),
called from KmtEnsureServiceState().
ROSTESTS-263
- Path must not contain whitespace characters.
- Path must be at least 2 characters long.
- Path must start with a backslash.
- Path must not end with a backslash.
- Path components must not end with a dot.
CORE-9529
- Improve the scrolling support for bitmaps, DIBs and text formats.
This completes the work started in CORE-10679 by Ricardo Hanke.
Includes scrolling with the keyboard and the mouse wheel.
- Add support for the CF_DSP* clipboard formats, as well as CF_TEXT
and CF_OEMTEXT.
- Add support for owner-display clipboard format CF_OWNERDISPLAY.
- Realize any palette found in the clipboard (CF_PALETTE) before
displaying the clipboard data format we want.
- Remove dead code.
- Update the file headers.
Before performing a read operation, FastFAT driver will
attempt to compute whether it would run out of stack
during the operation. If so, instead of attempting the
operation in the current thread, it will post the read
request to the overflow thread.
This should help with the regressions brought in by
94ead99e0c.
CORE-14601
Oneliner of the day... This typo just prevented the
whole feature to work properly. Because any allocated
work item would miserably fail to be freed.
This will obviously help real world FSD relying on
StackOverflow worker from FsRtl to work better!
CORE-14611
NETAPI32: Get rid of the old WINE NetServerGetInfo and replace it by a proper call to NetrServerGetInfo.
SRVSVC: Implement parts of NetrServerGetInfo.
srvsvc.idl: Hack around a bug in midl or rpcrt4. They are not able to handle a pointer to a union of pointers to structs. A pointer to a union of structs works a advertised.
- Make the reply buffer for IcmpSendEcho large enough to hold the reply,
even when testing a smaller size. This avoids a buffer overflow with ROS's
broken implementation.
- Avoid unnecessary initialization.
- Fix IcmpSendEcho return value check to succeed on Win2003.
- Don't free a string literal in the GetInterfaceName test.
In the lazy writer run, first post items that are queued for this.
Only then, start executing deferred writes if any.
If there were any, reschedule immediately a lazy writer run, to keep
Cc warm and to make it unqueue write faster in case of high IOs situation.
To make second lazy writer run happen faster, we keep our state active to
use short delay (1s) instead of standard idle (3s).
ScControlService:
- SEH-protext the calls to service control handlers.
- Return ERROR_EXCEPTION_IN_SERVICE if a service control handler throws an exception.
- Return ERROR_SERVICE_CANNOT_ACCEPT_CTRL if no service control handler has been registered for the service.
- RI_ScSetServiceBitsA: Just call RI_ScSetServiceBitsW.
- RI_ScSetServiceBitsW: Store the service bits in the service list entry.
TODO: Merge all service bits in a global variable and pass it to the server service. Maybe use netapi.I_NetServerSetServiceBits(Ex)?
Recent changes seem to show that it's not
required to be exclusive on VACB to be able
to flush it.
This commit goes with f2c44aa and fixes the
last issues going with copying huge files.
There are no longer BSODs (be it in Mm or Cc).
I could, with 750MB RAM extract a 2GB file from
a 53MB archive and copy a 2,5GB file from a VBox
share to the disk. Note that writes are often
deferred, so if copy works, it's not that fast for now.
Note that it also brings some beloved behavior from
Windows: copy times are totally unreliable now when
writes are deferred. Little remaining times when
actively copying, high remaining times when deferred
writes in action. And goes between both... Sorry! ;-)
https://xkcd.com/612/
CORE-9696
CORE-11175
Same mechanism exists in Windows (even their Cc
is way different from ours...) where when Cc is
out of memory (in their case, out of VACB), we
will start scavenge old & unused VACB to free
some of the memory.
It's useful in case we're operating we big files
operations, we may run out of memory where to map
VACB for them, so start to scavenge VACB to free
some of that memory.
With this, I am able to install Qt 4.8.6 with 2,5GB of RAM,
scavenging acting when needed!
CORE-12081
CORE-14582
Before any write operation that would involve caching, ask
the cache controler whether writing would make it exceed its memory
consumption. If so, queue the write operation for later execution.
In case the write operation can wait, then, the FSD operation will be
halted until the write is allowed.
I could test it successfully by copying huge files from a FAT volume to
another. The write is halted until some portions of the file is written
to the disk.
I could also properly install Qt (SDK) on ReactOS with this and less than 1GB RAM:
- https://www.heisspiter.net/~Pierre/rostests/Qt_OS.png
- https://www.heisspiter.net/~Pierre/rostests/Qt_OS2.png
CORE-12081
CORE-14582
CORE-14313
Adjusting refcount and enabling lazy-write for pinned
VACB makes it actually more efficient, often purging
data to disk, reducing memory stress for the system.
This is required for defering writes.
This commit unfortunately (?) reverts a previous revert.
CORE-12081
CORE-14582
CORE-14313
- Read the driver settings from the registry.
- Write the driver settings to the registry.
- Make the read-only edit control for the input buffer length look like a non-read-only one because it is controlled by the up-down control.
This is a PoC of what it's possible to realize thanks to an
already existing hack in ntoskrnl :-).
With this extension, on the kdb:> prompt, you're able to type
in commands that will be handled by the FastFAT driver and not
by the kernel, allowing internal debug, not possible otherwise.
So far, three commands exist:
- ?fat.vols: lists all the mounted volumes by FastFAT
- ?fat.files: lists all the files on a specific volume (with their attributes)
- ?fat.setdbgfile: allows watching on specifics files lifetime
This is obviously only the begin and could be greatly improved.
For instance, this is what allowed to debug CORE-14557
- Perform an immediate system shutdown if the timeout is zero, otherwise
display the countdown shutdown dialog.
- Cleanup pShutdownParams->pszMessage as soon as the dialog goes away.
- It actually appears (tested on Windows 2000 and 2003) that sending
WM_CLOSE messages from a user-mode app to the shutdown dialog really
do nothing (and in particular does not cancel the shutdown!), so modify
the code to take this fact into account.
This allows to break commands such as:
C:\ReactOS\system32> for %f in (*.*) do dir
as one would expect: stop the currently running 'dir' and the 'for'.
"bCtrlBreak" doesn't need to be volatile too.
Using CTRL-C to cancel command line input would leave the prompt in
a state where the next command would be ignored. For example:
dir<CTRL-C>
dir
would cause cmd.exe to ignore the second dir command.
CORE-11677
The standard Win32 Console Control Handler will give CTRL-C events to
processes spawned from cmd.exe. If cmd.exe calls GenerateConsolCtrlEvent()
then the child process will receive two CTRL-C events.
Once a directory is crossed (opened or a child is opened), associated
FCB structure is created in FastFAT, but also a stream FO for caching.
Up to now, due to an extra reference taken by the stream file object,
even when the directory was no longer used, the directory was kept in
memory: the FCB was never deleted, the file object was never dereferenced,
and the cache never released.
The immediate effect of this bug is that our FAT driver was leaking every
directory that was used affecting the whole OS situation. In case of
directories intensive operation (like extraction the ReactOS source code
in ReactOS ;-)), we were just killin the whole OS RAM without any way to
release it and recover.
The other side effects: IOs were faster as half of the FS was always
permanant in RAM.
This commit fixes the issue by forcing the FSD to release the FO,
and the cache when a directory is no longer used, leading to its
destruction in RAM.
Downside: on IO intensive operation, expect slowdowns, obviously,
there's less caching now. But more efficient!
CORE-14557
When no name is set in the file object, try to read the name
from the FCB. We only support FastFAT (ours) FCB for now.
This is clearly a hack, but for a kdbg command, so ;-)
It seems that on process killing, some VACB may be deleted while
still mapped. With current reference counting, they will actually
not be deleted, but leaked, and an ASSERT will be triggered.
CORE-14578
- Check the denominator for zero for both zoomed width and height.
- Do not allow to move zoom slider above possible position.
- Move commonly used `zoomTo` function into new header file.
CORE-14539
CORE-13447 CORE-14422
This fixes closing nLite 1.4.9.3 (.NET2) application, as well as any
other program that expects that the interactive window station is
reported as being visible. I thank Mark Jansen and Joachim Henze for
help and testing.
- Allow the driver to be unloaded, as on Windows.
- Use a static fast IO dispatch table, instead of allocating one and
risking any potential memory allocation failure.
- Update the file header.
- For non-PnP devices reported to the PnP manager through the
IoReportDetectedDevice() function, store the corresponding
service/driver name and (non-)legacy information inside their
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Enum\Root\ entries.
- Drivers flagged as "DRVO_BUILTIN_DRIVER" (basically, only those
created via a IoCreateDriver() call) have their "Service" name that
contain "\Driver\", which should be stripped before being used in
building e.g. the corresponding "DETECTEDxxx" PnP compatible IDs.
CORE-14247
- Use explicit REG_OPTION_NON_VOLATILE flag where needed in the
IopCreateDeviceKeyPath() calls.
- Save NULL-terminated REG-SZ string properties in the enumeration tree
for each device enumerated inside \Enum\Root\.
- Always use upcased key name for the "LEGACY_***" elements in \Enum\Root\.
- Add a default "ConfigFlags" value for the legacy elements.
- Simplify few parts of code.
The size is in bytes, not in pages! On x86 we got away with it, since PEB and TEB require only a single page and the 1 passed to MiInsertVadEx() was aligned up to PAGE_SIZE. On x64 this doesn't work, since the size is 2 pages.
Using static numbers was a bad idea in the first place, since we already deal with x64, which has different ordinals. Also @ will automatically do the same thing! It does not mean "no ordinal", it only means automatically apply incrementing numbers.
- "warning: comparison of unsigned expression < 0 is always false [-Wtautological-unsigned-zero-compare]"
- "warning: logical not is only applied to the left hand side of this comparison [-Wlogical-not-parentheses]"
CORE-14306
CORE-11637, CORE-12269
- Implement support for remote connections to NT computers to view their
event logs (and the event logs settings). For the moment the remote
computer name specification is is only available via the command-line.
Note that the paths to the event log files need also to be translated
to network-share format.
- Implement loading and saving the settings of the event viewer.
- Implement showing/hiding the events description pane, the ListView
grid, and whether or not to save the EventViewer settings.
- Handle hypertext link click handling for the RichEdit edit box of the
events description dialog.
This has have several benefits for ReactOS Cc:
- It helps reducing potential deadlocks situations in Cc
- It speeds up ReactOS by reducing locks
- It gets us a bit closer to Windows VACB
CORE-14349
The avoids race conditions where attempts to read from disk to
not fully initialized VACB were performed.
Also, added more debug prints in such situations.
CORE-14349
Doing this is not only wrong because it acquires the same spinlock twice,
it also completely breaks the TLB flushing logic in MiMapPageInHyperSpace.
If the PTE with Offset 1 is still valid when a wrap-around to 0 happens,
the TLB flush on wrap-around will not clear the entry for this previous page.
After another loop around all hyperspace pages, page 1 is re-used but its
TLB entry has not been flushed, which may result into incorrect translation.
This makes the function return the error instead of continuing and
performing use-after-frees operations.
This is likely to be a forgotten return!
CID 1434254, 1434268
Because we can encounter pending interrupts repeatedly,
HalpEndSoftwareInterrupt and HalEndSystemInterrupt already clean up the stack
space for their arguments (done for CORE-11123/CORE-14076).
However these functions are called from C functions such as KiInterruptDispatch
and HalpDispatchInterrupt2[ndEntry]. These callers also use up stack space,
and it is unknown how much.
To fix this, we simply reset the stack pointer to the location of the trap
frame, which is where it points during a first-level dispatch. This cleans
up the stack usage of any callers higher up, and is okay because a return
will happen through the trap frame anyway.
Dedicated to Pierre.
- Don't hardcode a buffer length in ExpandEnvironmentStringsW() call.
- If no file name is associated to a log (ErrorLog->FileName == NULL),
don't try to attempt looking at its file properties. This also allows
avoiding a crash in the FindFirstFileW() call under certain conditions
on Windows.
Variable 'Status' would be set but not used otherwise in this case.
This commit will not fix or improve CORE-13278,
but was spotted when having a look at that.
CORE-12269
- Add support for '/?' and '/l:' switches. The former displays usage
information. The latter allows to specify an event log file to load.
Also one can specify the computer name from which one wants to
retrieve events for display (work in progress).
- In EnumEventsThread(), move around the code that updates the
application title and status bar, so that the selected event log name
and loading status is always displayed first, while loading itself is
in progress.
- Initialize hwndMainWindow as it can be used before it is properly
initialized with an existing window handle.
- Transform ShowLastWin32Error() to a more generic ShowWin32Error().
- Use iswdigit() instead of isdigit().
- Add extra heap buffer cleanup.
- Split OpenUserEventLog() in two: OpenUserEventLog() that displays
an open-file dialog, and OpenUserEventLogFile() that does the real
work of allocating internal structures and opening the log file using
the event-log service. We also check that the file really exists
before proceeding further.
- Force treeview repaint when *children* items are inserted, because
otherwise the expand/collapse buttons do not display next to parent
items (this problem happens on Windows too).
- Reshuffle the resource string IDs.
This nice patch of Jerome Gardou improves palette
when VBEMP & inbuilt Mesa is used for
Diablo II, MS Age Of Empires 1, Anno 1602
to the state we had in SVN r75060 and below.
We already committed this fix into 0.4.8-release before.
Instead of loading systemcompatible.manifest as the implicit activation context, load forwardcompatible.manifest
Add a new assembly containing all apisets called ReactOS.Apisets and make it a dependency to forwardcompatible.manifest
- Use a more descriptive "unknown hard error" string than Windows' one.
- Improve the computation of the buffer size for the hard error message
by using _vscwprintf() instead of just "guessing" which size the fully
printf'ed message could be.
- In UserpCaptureStringParameters(), always capture the parameter
strings with a terminating NULL. Also, no need to convert them to
ANSI if they are used with a STATUS_SERVICE_NOTIFICATION message.
- Determine within UserpFormatMessages() the message box flags to be
used, either from the message parameters (if STATUS_SERVICE_NOTIFICATION)
or from the proposed response options and message status severity.
These flags are then used by both UserpShowInformationBalloon()/UserpShellHardError()
and by UserpMessageBox().
- Improve Message validation (especially for STATUS_SERVICE_NOTIFICATION).
- Try to display the hard error status number if it's an unknown one.
- Make UserpFormatMessages() always succeed by falling back on
static string buffers if allocating large string buffers fail due to
lack of memory. Then the hard error popups will have truncated
strings, but at least something will be displayed instead of nothing,
thus alerting the user that something went wrong anyway.
- Simplify a bit the code by moving the NtOpenProcess(),
UserpCaptureStringParameters() and UserpFreeStringParameters() calls
inside UserpFormatMessages(). This also allows simplifying its
prototype.
- Cache localized hard error message components: the status severity,
the system process name and the unhandled exception debug strings.
- Try not to fail too much. In particular, don't fail if a string could
not be correctly captured in UserpCaptureStringParameters() (we then
use a default empty string). The next aim is to make UserpFormatMessages()
not failing at all.
- Use RtlStringCbPrintf(Ex)W where possible so that one can use counted
(and not NULL-terminated) UNICODE_STRINGs where possible. This allows
using counted resource strings without having to allocate memory.
- If available, prepend the window title of the application that
triggered the hard error to the hard error message box caption.
In particular do not always fail as soon as there is an error, because they may be the sign of an OS problem and this is precisely in this case that we want to display a hard-error.
- Use hard-error captions depending on the status code severity.
- Improve handling of the STATUS_UNHANDLED_EXCEPTION error. Start
implementing the display of the dialog button descriptions ("OK" to
terminate the application; "Cancel" to debug the application).
- Add support for the "ErrorMode" and "ShellErrorMode" registry values
under HKLM\System\CurrentControlSet\Control\Windows.
- Use MessageBoxTimeoutW() in UserpMessageBox() to support different
message box timeouts.
- Log hard errors into the System event log.
- Make the "Press Crtl-Alt-Delete..." text control as wide as possible so that the text will fit in one line for all languages.
- Make the albaninan SAS dialog look like the SAS dialog for the other languages.
With these, we generate the WM_CONTEXTMENU and NIN_(KEY)SELECT
shell icon notifications that applications expect when they handle
shell notification icons with uVersion >= 3.
This fixes in particular the previously unresponsive icon of KVIrc 4.x,
and more generally *all* the notifiation icons of Qt applications.
CORE-10605 #resolve
This replaces old link to svn.reactos.org with rapps.reactos.org
which is a new stable location for RAPPS database and future community
website. That said, svn.reactos.org link will be kept for some time.
TODO: make this a configurable value.
CORE-14351
- Use the "%d days" format for timeouts longer than a day.
- Fail if timeout is 10 years or longer.
- TODO: Replace format strings by resources. German WinXP uses "%d days" instead of "%d Tage". We can do better! ;-)
* [SERVMAN] Italian update translation
* [SERVMAN] Small grammatical typo fixed
* [SERVMAN] Word replaced
- "Accedi" would be more convenient than "Accedere"
- Replace the winlogon icon by the warning icon.
- Add more space for the main text.
- Move the shutdown time into a separate line to ensure it is always fully visible.
- Add more space for the shutdown message and keep the message empty by default.
- Clean up the resource IDs.
- Replace the UNICODE_STRING usMessage by a PWSTR pszMessage.
- Use the "%02d:%02d:%02d" time format and get rid of the safe string printf because the string will NEVER be longer than 8 characters.
- Rename the timer id constant because it is NOT a resource id.
- Rename variables according to the coding style.
- Add empty lines to separate logical blocks of code.
- Add spaces according to our coding style.
- Move g_hShutdownDialog and g_bShuttingDown into the shutdown parameters struct.
- Pass a pointer to the shutdown parameters to the shutdown thread and the shutdown dialog and use the pointer instead of the global variables.
- Move the timer code into a separate function because it requires local variables.
- Fix a copypasta.
- Remove dead code.
Thanks to mudhead, who spotted most of these.
NB:
This EXPERIMENTAL_MOUSE_CURSOR_SUPPORT code isn't compiled by default since r7865.
CORE-14511
- Small Italian translation update;
- Translation update for Romanian;
- Small update (just two strings translated based on the latest RAPPS commits);
- Header E-mail changed.
Import the following Wine commits:
* ea2798f1ce9 Iván Matellanes: msvcrt: _setmode should check if file descriptor is valid.
* 9b495caacfd Carlo Bramini: msvcrt: Call SetFilePointer() with correct parameters.
* 13f34481b26 Carlo Bramini: msvcrt: Use MSVCRT_xxx macros instead of the same ones from GLIBC.
* 781b069ed81 Piotr Caban: msvcrt: Don't close HANDLE if stdout or stderr is sharing it.
Import Wine commits by Piotr Caban:
* 6ed69c107f8 msvcrt: Avoid using global critical section while allocating new file descriptors.
* 725e4733cf8 msvcrt: Remove no longer needed global lock from functions using fd critical sections.
Import Wine commits by Piotr Caban:
* 417b6056877 msvcrt: Fix possible deadlock in dup2 function.
* e7fc2370885 msvcrt: Don't use fdend in create_io_inherit_block.
Import Wine commits by Piotr Caban:
* c7b0fb51543 msvcrt: Use fd critical section in _dup2.
* b052afd4afa msvcrt: Use fd critical section in msvcrt_free_fd.
* 16285f5700f msvcrt: Use fd critical section in read.
* 9ce2a8a60c1 msvcrt: Use fd critical section in create_io_inherit_block.
* f00b7f40584 msvcrt: Use fd critical section in _setmode.
* ba757eaec74 msvcrt: Use fd critical section in _write.
* eec7a565bec msvcrt: Don't lock __badioinfo.
Import Wine commits by Piotr Caban:
* 6c2d4f1092d msvcrt: Use fd critical section in _fstat64.
* 9278190d468 msvcrt: Use fd critical section in _futime64.
Import Wine commits by Piotr Caban:
* 4508e665c07 msvcrt: Use fd critical section in _commit.
* afc6962a24a msvcrt: Use fd critical section in _close.
* 9a4cfbc3edc msvcrt: Use fd critical section in _eof.
* ffdd16cc0d7 msvcrt: Use fd critical section in _locking.
* fa6de597700 msvcrt: Use fd critical section in _lseeki64.
* f2f45d5fe67 msvcrt: Use fd critical section in _chsize_s.
- Use a stop event to terminate the dispatch loop.
- Report to the services manager that the service can be stopped or shut-down while it is running.
CORE-14390
This allows being consistent between newly created and looked up
so that VACB can always safely be released.
Should really help with reference issues.
CORE-14481
CORE-14480
CORE-14482
* [BROWSEUI] Update German translation
* [SHELL32] Update German translation
* [COMCTL32] Update German translation
* [COMDLG32] Update German Translation
* [SETUPAPI] Update German translation
* [EXPLORER] Update German translation
This fixes various bugs linked to VACB counting:
- VACB not released when it should be
- Reference count expectations not being accurate
For the record, VACB should always have at least a reference
count of 1, unless you are to free it and removed it from
any linked list.
This commit also adds a bunch of asserts that should
help triggering invalid reference counting.
It should also fix numerous ASSERT currently triggered and
may help fixing random behaviours in Cc.
CORE-14285
CORE-14401
CORE-14293
FDI allows to have user-defined callbacks for file handling.
Since it doesn't provide support for Unicode we convert strings to
multi-byte UTF-8 and handle them appropriately in the callbacks. They
are properly null-terminated so FDI won't choke when doing operations
with strings.
Thanks to hbelusca and mjansen for the help.
CORE-14466
- Fix the layout of the normal master dialog layout.
- Change the name of the normal and tray master dialogs.
- Add the WS_EX_LAYOUTRTL style to the hebrew dialogs.
- Add the small master dialog.
- Add support for tray (/t), small (/s) and normal (/n) dialog sizes.
- Fix the font size calculation for the dialogs.
- Fix the dialog unit to pixel conversion.
- Fix the client window size calculation.
- Do not leak font handles.
* Convert to wWinMain()
* Add a resource file
* Use a slightly friendlier application name
* Add code to load the output path from the Registry
* Add localized string resources
"warning: argument to 'sizeof' in 'memcmp' call is the same pointer type 'SDBQUERYRESULT_VISTA *' (aka 'struct tagSDBQUERYRESULT_VISTA *') as the destination; expected 'SDBQUERYRESULT_VISTA' (aka 'struct tagSDBQUERYRESULT_VISTA') or an explicit length [-Wsizeof-pointer-memaccess]"
CORE-14306
A bit of history: in r71528, I tried to fix our explorer often
crashing while browsing directories. It was linked to the fact
that a notification result may arrive while the notification
structure had already been deleted.
The fix for this was actually broken and was leading to a double
leak: the notification structure was leaked. But also the handle
to the directory that had been browsed!
This means that the directory couldn't be modified anymore as
a leaked handle to it was still open.
Actually, when notifications are cancel, the kernel properly
calls the notification routine, but with a specific error code.
So the correct fix is to stop handling that notification when
we receive this error code. This is the correct fix with no leaks.
This commit is a complete r71528 revert with the appropriate fix.
CORE-10941
CORE-12843
This is a temporary fix for a timeout in windowscodecs:tiffformat.
The test ends up invoking Win32ErrorHandler with:
_TIFFVSetField: <IStream object>: Bad value 0 for "ResolutionUnit" tag.
Ideally we'd keep the MessageBox version of the error/warning handlers but
make them respect the current error mode to avoid blocking tests and other
non-interactive applications.
The initial tests in send.c validate correct behavior of send/sendto on
disconnected sockets (CORE-9810), fixed in r68129.
However, the helper functions are generic, so they can be used for additional
tests against AFD. Because AFD's create packet structure changes between
Windows versions, the functions check the OS version to determine the right
layout.
Tests succeed on Win2003 as well as Win10.
- Make sure the DWLP_* values are correct on _WIN64
- Don't use the DWL_* constants, they are not portable. Enforce this by removing them entirely from winuser.h
- Make sure Get/SetWindowLong*Ptr* is used and pointers are not truncated to LONG
Previously, we would keep sampling the CPU frequency until two subsequent
samples differed by at most 1 MHz. This could take several seconds, and would
unnecessarily delay boot.
Instead, if sampling is too unreliable, just give up and calculate the average
frequency from 10 samples. This is no worse than picking the frequency that
just happened to be returned twice in a row.
The fact that this method of sampling fails could indicate that there's a
problem with our performance counter implementation or timer interrupt,
but that's a separate issue...
- Use a new work_available event to allow ProcessPlayingNotes to sleep when no
notes are to be played.
- Get rid of the pointless thread_termination_complete event, wait on the
thread handle instead.
- Don't leak thread_handle.
This avoids locking Cc for too long by trying to read-ahead data which
is already in cache.
We now will only schedule a read ahead if next read should bring us
to a new VACB (perhaps not in cache).
This notably fixes Inkscape setup which was slown down by read-ahead
due to continous 1 byte reads.
Thanks to Thomas for his help on this issue.
CORE-14395
- Volume, Power and Hotplug icon: Use timers to distinguish between single and double clicks.
- Volume icon: Show the tray volume dialog on a single click.
SendNotifyMessageW: It doesn't support messages with pointers so there is no need to call MsgiUMToKMMessage
SendMessageTimeoutW, SendMessageTimeoutA: These two do support marshaling pointers so they need to use MsgiUMToKMMessage.
This is actually a bug that happens only in the rare case where we send a WM_COPYDATA with a timeout.
- Set some last errors.
- Fix error code returned by ScLookupServiceByServiceName().
- Check the validity of the handler proc in RegisterServiceCtrlHandler(Ex)W().
- Improve some traces; comment some code.
The reserve IRP is an IRP which is allocated on system boot and kept during
the whole system life. Its purpose is to allow page reads in case of
low-memory situations where the system doesn't have enough memory left
to allocate an IRP to read from the page file (would be catastrophic situation).
services\database.c:
- Refactor ScmControlService() so that it can be used to send the dispatcher loop stop command.
- Separate the code to decrement the image run counter from the service image cleanup code.
services\rpcserver.c:
- RSetServiceStatus(): Stop the dispatcher loop when the image run counter is zero and remove the service image after that.
advapi32\service\sctrl.c:
- Do not terminate the service dispatcher loop when the last service is being stopped. Wait for an explicit dispatcher stop command (empty service name).
CORE-12413
(So the fun begins)
In spite of what VFATLIB headers pretend, there's not magic in FAT boot sector.
The 3 first bytes are just the jump instruction (to the boot code). No jump, no boot.
Also, some (many?) FAT implementations rely on the jump code to help detecting that
a FAT volume is really a FAT volume. Like MS FastFAT. Or our own FAT recognizer in FS_REC.
The story is that, up to that commit, we zeroed the 3 first bytes; leading to broken
FAT volumes.
This got hidden in most cases by the fact that during setup, when we install boot
loader, we erase parts of the boot sector, including the jump instruction, making the
volume valid again. But that wouldn't fix secondary volumes where the boot loader isn't
installed.
And, also, imagine a scenario where you want to install ReactOS on a newly formatted volume
with MS FastFAT instead of our own implementation... That would simply not work to
the fact that the driver wouldn't recognize the fresh formatted volume!
(So the non fashion begins)
Fix this by putting a not that valid jump into the boot sector when formatting our
partitions. That way, our volume is always regarding a FAT view point. But, instead of
putting values that mean (nearly) nothing. We should also put a dummy bootloader
displaying the user and error message, as done by dosfstools.
(So the hope begins)
This opens the way for trying to install ReactOS with MS FastFAT (doesn't work yet).
CORE-11819
CORE-14362
The two methods were completely broken because they passed a full pidl to the parent folder (and not desktop folder).
Compile SHGetNameFromIDList to use it in CShellItem but don't export it.
on 32bit builds CSIDL_PROGRAM_FILESX86 is treated as CSIDL_PROGRAM_FILES and on wow64 builds CSIDL_PROGRAM_FILES gets treated as CSIDL_PROGRAM_FILESX86
- ScmUnloadDriver: Handle failed unload properly.
- ScmGetDriverStatus: A driver is running when its driver object was found. Otherwise it has been stopped.
- ScmControlDriver: Check a drivers status before and after unloading.
CORE-14317
- We should not open a new connection to request a certificate.
- Update the issuer / subject for the LE certificate.
- User proper types.
- Require all fields that we check to be present.
Without checking the public key or the entire certificate it's still not secure, but we are a step closer.
Dedicated to Joachim Henze
CORE-14350
- In HKLM\SYSTEM\CurrentControlSet\Control\Windows, add "Directory"
and "SystemDirectory" values, giving the ReactOS installation
directories (purely informative).
- Also add the "ErrorMode" value that specifies how the hard-error
popups are displayed. This value will be used later.
- Fix event log sources ordering.
Set the view object as the site of the drop target.
Implement GetItemPosition, SelectAndPositionItems, IsDropOnSource, GetDragPoint.
Use DROPEFFECT_COPY only when the item supports it.
This patch reduces failures of SHGetFileInfo function. CORE-7159
* improve WideByteToWideChar calls
* fix bugs related to file attributes and SHGFI_EXETYPE
* SHGFI_USEFILEATTRIBUTES and SHGFI_ICON fix
* s/sizeof(temppsfi)/0/
CDefView: Make GetItemObject return an IContectMenu after setting itself as the site.
CDefViewBckgrndMenu: Make the inner context menu use the same site. Implement handling CMDSTR_VIEWLISTA and CMDSTR_VIEWDETAILSA.
CDefaultContextMenu: CMDSTR_VIEWLISTA and CMDSTR_VIEWDETAILSA shouldn't be handled in this class.
Fixes CORE-14327 and the two buttons that change the view mode
It shouldn't be necessary for QueryContextMenu to be called before InvokeCommand.
Move retrieving the IContextMenu of the folder to the initialization of the class.
CORE-12866
[SHLWAPI] Refactor the SHCreateWorkerWindowW() prototype to match its ANSI SHCreateWorkerWindowA() counterpart.
The last parameter is really to be understood as an extra window data, and not a "message result" (as it would be the case for dialog window procedure).
That is why I also remove the mention of "DWLP_MSGRESULT" in the SetWindowLongPtrW() call.
SHCreateWorkerWindowA() had it OK but SHCreateWorkerWindowW() did not.
------------------
[SHLWAPI] Make SHCreateWorkerWindowA() and SHCreateWorkerWindowW() x64-compatible.
The first parameter of these functions is a pointer to a window procedure, having a definite prototype, so employ a correct typedef WNDPROC,
which ensures both correct pointer size and parameter type enforcement.
This also ensures that we use instead a correct pointer size, since otherwise LONG remains 32-bits for Windows compatibility on x64 platforms.
The wndProc parameter is thus casted to LONG_PTR to comply with the SetWindowLongPtrA/W calls.
In SHCreateWorkerWindowW(), the last "wnd_extra" parameter should also be LONG_PTR to be able to pass 64-bit data pointer on x64 platforms.
Therefore fix also setting the wc.cbWndExtra size. One should note that the ANSI SHCreateWorkerWindowA() function had everything OK already.
is set before locking is actually attempted for Cc worker thread (lazy write/read ahead).
So in case locking fails, the top level IRP is set and never unset, and latter
call will hit the assert where it expects the top level IRP to be NULL.
Should be reported upstream (tm).
CORE-14315
- The initial dwWin32ExitCode for a disabled service is ERROR_SERVICE_DISABLED and ERROR_SRVICE_NEVER_STARTED for non-disabled services.
- The initial dwWaitHint is 2000 (2 seconds) for (user-mode) services and 0 for drivers.
- Move all driver-related code to driver.c.
WatchList is a set of array objects, initialized with "new[]", so it should be freed with "delete[]" to free all of its elements. Otherwise using only "delete" only frees the first variable but not its array. This would lead to an undefined behaviour.
Standard shared cache map provides space for a private cache map, do the same
and make it available for the first handle. It avoids two allocations in a row.
This halfplements CcScheduleReadAhead() which is responsible for finding the next reads
to perform given last read and previous reads. I made it very basic for now, at least
to test the whole process.
This also introduces the CcExpressWorkQueue in the lazy writer which is responsible
for dealing with read ahead items and which is dealt with before the regular queue.
In CcCopyData(), if read was fine, schedule read ahead so that it can happen in background
without the FSD to notice it! Also, update the read history so that scheduling as a
bit of data.
Implement (à la "old Cc" ;-)) CcPerformReadAhead() which is responsible for performing
the read. It's only to be called by the worker thread.
Side note on the modifications done in CcRosReleaseFileCache(). Private cache map
is tied to a handle. If it goes away, private cache map gets deleted. Read ahead
can run after the handle was closed (and thus, private cache map deleted), so
it is mandatory to always lock the master lock before accessing the structure in
read ahead or before deleting it in CcRosReleaseFileCache(). Otherwise, you'll
just break everything. You've been warned!
This commit also partly reverts f8b5d27.
CORE-14312
before the call to CcRosReleaseFileCache() which expects to have it to properly clean the file.
So, move deletion code to CcRosReleaseFileCache() so that he's the only one to handle private map.
Should hopefully fix all the recent buildbots issues (and the universe perhaps, who knows?)
This reverts BCB being lazy written when marked dirty.
We'll go back to this behavior when this part will have been reworked and stabilized.
CORE-14263
CORE-14279
CORE-14285
This is better than wasting space on null chars (and having Git mangle it).
Also ensure that it gets CRLF newlines since its contents go in the binary.
We get rid of the old iLazyWriterNotify event in favor of work items
that contain an event that lazy writer will set once its done.
To implement this, we rely on the newly introduced CcPostTickWorkQueue work queue
that will contain work items that are to be queued once lazy writer is done.
Move the CcWaitForCurrentLazyWriterActivity() implementation to the
lazy writer file, and reimplemented it using the new support mechanisms
Instead move to a threading model like the Windows one.
We'll queue several work items to be executed in a system thread (Cc worker)
when there are VACB that have been marked as dirty. Furthermore, some delay
will be observed before action to avoid killing the system with IOs.
This new threading model opens way for read ahead and write behind implementation.
Also, moved the initialization of the lazy writer to CcInitializeCacheManager()
it has nothing to do with views and shouldn't be initialized there.
Also, moved the lazy writer implementation to its own file.
Modified CcDeferWrite() and CcRosMarkDirtyVacb() to take into account the new threading model.
Introduced new functions:
- CcPostWorkQueue(): post an item to be handled by Cc worker and spawn a worker if required
- CcScanDpc(): called after some time (not to have lazy writer always running) to queue a lazy scan
- CcLazyWriteScan(): the lazy writer we used to have
- CcScheduleLazyWriteScan(): function to call when you want to start a lazy writer run. It will make a DPC after some time and queue execution
- CcWorkerThread(): the worker thread that will handle lazy write, read ahead, and so on
Wine's comctl32 somehow fails to load an 24bpp image with ImageList_LoadImageW
when crMask is CLR_NONE so use a color that doesn't exist in the bitmaps until
comctl32 gets fixed.
I prefer to add this silly hack in a non critical dll than make a hasty change in the
critical comctl32 until some proper tests are written for this absurd failure.
Such a manifest is meant to be activated by the caller of the exports of the dll.
This means that the user of syssetup which is setup.exe is the one that should do the proper handling of the manifests.
Since setup.exe is the one that needs to do the right thing, add a CREATEPROCESS_MANIFEST_RESOURCE_ID to it and avoid
dealing with ISOLATIONAWARE_MANIFEST_RESOURCE_ID altogether (which is supposed to be used by activating the activation context
before calling the exports of the dll that has it). We don't really use ISOLATIONAWARE_MANIFEST_RESOURCE_ID correctly in any
place in the tree so let's rely on CREATEPROCESS_MANIFEST_RESOURCE_ID for now.
This fixes one part of CORE-14228 where no image was shown in the theme selector buttons. This happened because comctl32 was
statically linked and calling ImageList_LoadImage ended up creating an image list with comctl32 v5. Then we passed this image
list to a button created with comctl32 v6 and as a result the image list was unusable. This is yet another issue that needs
to be addressed in the future. Namely, it seems that in windows comctl32 v6 can use image lists created by comctl32 v5 whereas
we cannot.
This fix is not a hack but cleverly works around these two bugs that need to be addressed some time in the future.
CORE-14228
To reflect recent changes in taskbar height, icon spacing,
clock x and y margins.
Also make sure all shots show the exact same time
and choose a more localization agnostic time.
This means that MmSystemCacheStart, MmSystemCacheEnd, MmSizeOfSystemCacheInPages
have now a valid value.
System cache is not used atm the moment though. MmMapViewInSystemCache() is to
be implemented, and Cc is to be made aware of this.
CORE-14259
[FONTEXT]: At the moment there is a comment in the en-US.rc resource file that advises to not translate it. I only created this file as a basis for the future. The file will be updated later when there is the possibility for FONTEXT to be translated further.
- Change MM_SYSTEM_SPACE_START to 0xFFFFF88000000000
- Move MI_DEBUG_MAPPING to the end of the system PTE range
- Add MI_SYSTEM_CACHE_START and MI_SYSTEM_CACHE_END, which is in the range that Vista uses as dynamic VA space for cache and other allocations
- Wrap x86 specific code that makes now invalid assumptions about the address space layout in #ifdef _M_IX86
[ROSLOAD]: Stubplement OslpCheckForcedFailure, OslpGetSetBootStatusData, OslSetBootStatusData, OslGetBootStatusData.
[ROSLOAD]: Stub OslpInitializeBootStatusDataLog, OslpReadWriteBootStatusData.
[BOOTLIB]: Fix BlAppendBootOptionString to accept an actual BCD ID instead of hardcoding LibraryPath.
[BOOTLIB]: Fix BlAppendBootOptionBoolean to accept an actual BOOLEAN value instead of hardcoding TRUE.
[BOOTLIB]: Implement BlDeviceIsVirtualPartitionDevice.
[BOOTLIB]: Add missing BcdOSLoaderInteger_ForceFailure BCD value. Add BCDE_OSLOADER_TYPE_BOOT_STATUS_POLICY based on BcdEdit.exe and Geoff Chappel site.
[NDK]: Document latest RTL_BSD_DATA as of RS3 based on ext.dll WinDBG
extension "!blackboxbsd" which outputs the entire structure field by
field :).
[NDK]: Update RTL_BSD_ITEM_TYPE thanks to ole32 symbols
[RTL]: Move system volume APIs to sysvol.c
[RTL]: Fill out BsdItemTable based on actual field offsets/sizes and not
hardcoded magic numbers which we won't ask where they came from.
[RTL]: Make RtlCreateBootStatusDataFile use an appropriate structure for
initializing the buffer instead of "UCHAR Buffer[12] = {0xC,0,0,0,
1,0,0,0, 1, 0x1e, 1, 0};" which appears like magic knowledge.
[RTL]: Rename "WriteMode" to "Read" in RtlGetSetBootStatusData since
it's much less confusing.
[RTL]: Some formatting fixes, SAL updates.
- Introduce the TRAYNOTIFYDATAW structure, as documented by Geoff
Chappell in "WM_COPYDATA for Taskbar Interface", at
http://www.geoffchappell.com/studies/windows/shell/shell32/api/shlnot/copydata.htm
that is the data structure passed between shell32 and explorer for
communicating shell notify icon information.
- In Shell_NotifyIcon(), correctly capture the (ANSI and) UNICODE
structures provided by the caller, properly taking into account for
the different NOTIFYICONDATA structure sizes existing out there.
The different strings are now properly null-terminated (especially
szTip if it needs to be truncated out), and the flags validated.
- Remove the now unneeded "SHELL_NotifyIcon()" helper function.
[EXPLORER] Use TRAYNOTIFYDATAW and adjust the callers.
[ROSLOAD]: Stubs for OslFatalErrorEx, OslAbortBoot.
OslpSanitizeLoadOptionsString, OslpSanitizeStringOptions
[ROSLOAD]: Implement OslBlStatusErrorHandler,
OslpRemoveInternalApplicationOptions
[ROSLOAD]: Begin work on OslPrepareTarget. Small misc. fixes.
[BOOTLIB]: Extend BlStatusError to handle a custom status error handler.
[ROSLOAD]: Implement OslExecuteTransition, OslArchKernelSetup,
ArchRestoreProcessorFeatures, OslArchTransferToKernel.
[ROSLOAD]: Factor out the logo code. Clean ups.
- Start to doxygenate the library, focusing in great details on the
functions of the "outstream" module.
- Add a K32LoadStringEx function that expands (K32)LoadString by
allowing a LanguageId parameter to be able to load strings from other
languages than the current one.
- Add "ConResMsg*" helper functions to be able to (format and) print
message strings with inserts that come *NOT* from a message table (as
usual) *BUT* from resource string tables.
Will be helpful for CORE-14265 in particular.
[CMD] Fix the call to ConMsgPrintfV().
- Add missing error code to DDK/NDIS
- Fix error code in IP driver
- Patch MSAFD to reply correct WSA error code
- Fix IcmpSendEcho function
- Fix returning error code in ICMP_ECHO_REPLY (see MSDN)
- Fix returning host address (using GetBestInterface and GetIpAddrTable functions)
- Fix GetBestRoute function (it is used by GetBestInterface)
Relates to #318 and CORE-14241
-VSSolution switch now outputs to a dedicated folder if running from source directory.
-Give an error if the user is trying to mix ninja and VSSolution in the same output directory.
-In a MingW environment, give an error on VSSolution switch and give a warning on RTC switch.
-Fix -DNEW_STYLE_BUILD=0 option.
-Ensure the source path doesn't contain any spaces.
-Issue a warning if passed an invalid switch.
-Don't announce completion if there were errors.
-Make end message specific for VSSolution.
-Replace redundant exit commands with quit label, as suggested by HBelusca.
As code grows, this will allow for better maintenance of the console
stream code. In particular the input stream module will contain special
code for handling TTYs, and this is something not all console programs
will need. Having this code in a separate module will allow for the linker
to possibly remove this code when it is unused.
MSDN says that these values are not used but apparently they are used as a padding around the buttons.
This helps to center the buttons in the notification area.
[ROSLOAD]: The IA32_MISC_ENABLE MSR bit actually _disables_ NX, so
diabling the bit _enables_ NX.
[ROSLOAD]: The MSR_XD_ENABLE_MASK is for the _high_ DWORD of the MSR
(bit 34), not the low.
[ROSLOAD]: Fix printf.
[ENVIRON/ROSLOAD]: Switch to CPU_INFO structure instead of blindly using
INT[4]. Revealed a bug in the PAE detection which was reading out of
bounds.
- CcUnpinDataForThread() only release VACB when the last BCB reference is gone. This avoids having a valid BCB with an invalid VACB
- CcRosMarkDirtyVacb() will only accept non-dirty VACB now. This avoids a major bug where a an already dirty VACB was over-dereferenced
- Thanks to previous point, simplify CcRosUnmapVacb(), CcRosReleaseVacb() implementation
- And only set VACB dirty once in CcSetDirtyPinnedData()
- Add a few sanity checks
With that I can again install ReactOS with 128MB RAM :-).
CORE-14263
CORE-14268
- Plus: Update license header.
- Re-use PrintServiceStatus(), as SERVICE_STATUS_PROCESS is in fact an extension of SERVICE_STATUS.
- Use the "SERVICE_*" defines, not magic numbers.
- Use the tab title from the corresponding dialog resource caption string.
- Fix some translations, and complete the french one.
- Use DialogBoxParamW and MAKEINTRESOURCEW (with W).
Added Estonian translation for the common base applications Notepad and MS Paint. These are the initial translations, and are subject to change, as they may not be the most perfect.
- Host/net unreachable error should be obtained from EchoReply->Status
- Use GetNameInfoW to decode actual response addresses (borrowed from tracert code)
CORE-14241 #resolve
Namely, implement CcSetDirtyPageThreshold() and add support for it
in CcCanIWrite().
Also added my name in the headers of the few files I touched tonight.
CORE-14235
Namely, implement CcCanIWrite() (very basic, and likely wrong).
And implement CcDeferWrite() which will queue the write operation.
In CciLazyWriter() (which may be renamed CcWorkerThread() ;-)),
handle the queued write operations one by one. This is likely
not to be accurate, but, given we have only on FS supporting
this for now (NFS / RDBSS / Shares), this is OK.
CORE-14235
Experiment and MSDN tend to show that a dirty BCB is queued for lazy write.
This will do the job here!
Also, renamed CcRosMarkDirtyFile() which is more accurate, and added a new
function CcRosMarkDirtyVacb() which just takes a VACB as arg (expected locked)
and marks it dirty (using previous implementation). Make CcRosMarkDirtyFile()
use it.
CORE-14235
CcDirtyPageThreshold is not here to compute when you have to write,
but to know where you have to deny writes.
More commits to come in that direction!
CORE-14235
Apparently the indices provided in NMTBGETINFOTIP (TBN_GETINFOTIP data struct), are not reliable,
but since the lParam values are, and it saves us a lookup, we will be using those instead. Win-Win!
- uVersion will only be truly useful when Vista+'s V4 style notification icons are implemented.
- Balloon notifications do not yet support queuing and auto-closing.
- Force the notification icon tooltips to always show even if the taskbar isn't foreground.
[ROSCTRLS.H] Implement CTooltips class which manages a comctl32 tooltips window.
This removes the "modified page writer" thread in Mm that was regularly blindly
attempting to flush dirty pages to the disk.
Instead, this commit introduces a lazy writer that will monitor dirty pages count
and will flush them to disk when this count is above a threshold. The threshold is
computed on Cc init.
Compared to what was done previously, this lazy writer will only write down files
that are not marked as temporary.
The mechanisms involved in this lazy writer worker are well described in Windows
Internals 4th editions (constants are coming from it ;-)).
Also fixed a bad (and old!) bug in CcRosFlushDirtyPages() where target count could
be overflow and the function would spin forever while holding the VACBs lock. This is
mandatory as now lazy writer will call it with "random" values.
This also allows implementing CcWaitForCurrentLazyWriterActivity() :-).
Also renamed DirtyPageCount to its MS equivalent.
CORE-14235
This Wine sync (that was not done against WineStaging) erased code
that had been submitted to WineStaging by the ReactOS team (clist.c
fixes, see r73565 a05881c and r73976 7b6b597) or added in r68042 9a78054.
To that aim I add explicit #if(n)def __REACTOS__ guards around the code.
All it will do for now is dumping the MFT / Files and their attributes.
It lacks many features, could be really improved and is bugged.
Note that logical sectors dump seems to be broken. MS NFI and this one don't agree on values.
Developed against W2K3.
PDBs reveal that these functions are also used in winspool.drv, but not imported from spoolss.dll to retain the client/server architecture.
As winspool.drv highly benefits from the MarshallUp* functions, I put them in a source file shared between spoolss.dll and winspool.drv.
The added API Tests cover my testing and all implemented features.
One more item done from https://reactos.org/wiki/Printing !
This would affect reads/writes on large volumes where offset is higher than what a ULONG can hold.
This really nasty bug was hitting CcMapData() but also CcPinRead() (due to the nature of its implementation)
and both were returning garbage data under certain circumstances with Ext2Fsd.
This should (I hope!) help some other FSDs to work better in ROS.
CORE-12456
Print a debug message whenever an application newer than ReactOS' internal NT version is started.
This gets a simple "Hello World" compiled by VS' 2017 regular "v141" toolchain to run under ReactOS.
Please retry NT6+ applications!
Rename SetupDiSetDeviceRegistryPropertyW to IntSetupDiSetDeviceRegistryPropertyAW and use it for both (A/W).
The FIXME in the A-Version was already implemented in the W-Version.
CORE-14173.
As a bonus, the public database at https://svn.reactos.org/packages/rappmgr.cab is now automatically updated with every commit to that repository.
Thanks to Alexander Shaposhnikov for the idea and help!
This reverts ea4daf8b68 committed 2017-12-09 to master, the commit was named
"[WIN32K] Fix NtGdiGetTextFaceW to report face name correctly (#173)",
reverting got ack of khmz,
reopening less severe CORE-14071 accordingly
This is assorted with misc fixes to make this code closer to what we currently have in FastFAT.
This also allows fixing a memory leak in case of single entry return.
CORE-13367
Note: before we had a BOOLEAN parameter called StoreInstruction, but in reality it was not specifying whether the fault was from a store store instruction, but whether it was an access violation rather than a page-not-present fault. On x86 without PAE there are only 2 kinds of access violations: (1) Access of a kernel mode page from user mode, which is handled early and (2) access of a read-only (or COW) page with a writing instruction. Therefore we could get away with this, even though it relied on the wrong assumption that a fault, which was not a page-not-present-fault, was automatically a write access. This commit only changes one thing: we pass the full fault-code to MmAccessFault and handle the rest from there in exactly the same way as before. More changes are coming to make things clear.
Based on an original patch by Timo Kreuzer, with modifications by me to adapt it to latest HEAD and use a single exit path through the Cleanup label. This reliably frees all allocated handles.
The original code returns STATUS_SUCCESS for many cases. This has been preserved.
In the future, it should be checked though whether returning success is appropriate for all these cases.
CORE-6844
This avoids the open failing on certain file systems where GENERIC_WRITE access would be denied
for a readonly file.
This is an addendum to 82f44a2.
CORE-14158
On application crash, drwtsn32 will attach to the application and try to get a dump, consisting of:
- List of loaded modules
- List of loaded threads
- Per thread, a stacktrace
- Per thread, a small hexdump from the stack
- Per thread, a dump of the most common registers
This dump is saved to the desktop, and the user is notified of the dump being dropped there.
CORE-14180
#145
In such situation, try to drop the readonly attribute before overwritting the file.
This fixes setup not being able to overwrite FAT bootcode when reusing a FAT partition
already installed without reformatting.
CORE-14158
- Added a 'Selected for installation' category that shows what was selected.
Selection is now kept between categories.
- New string is added to resources.
ru-RU.rc and uk-UA.rc strings are translated.
CORE-13789
- Full rewrite. This commit replaces the old utility.
- Use the new ICMP APIs instead of manually crafting ping requests using raw sockets.
- Add support for additional languages (the previous utility was hardcoded)
- Add support for IPv6
- Make the icmpapi header C++ compatible. (we don't appear to sync this with wine anymore.)
- Now runs on Win10, is much more reliable, and brings the code somewhat into the 21st century.
(It's currently missing source routing (-j), but as most routers disable this anyway, I'm not sure that it's worth adding)
[REACTOS] Misc 64 bit fixes
* [NTOS:MM] Allow MEM_DOS_LIM in NtMapViewOfSection on x64 as well
* [NTOS:MM] Implement x64 version of MmIsDisabledPage
* [HAL] Remove obsolete code
* [NTOS:KE] Fix amd64 version of KeContextToTrapFrame and KeTrapFrameToContext
* [XDK] Fix CONTEXT_XSTATE definition
* [PCNET] Convert physical address types from pointers to PHYSICAL_ADDRESS
CORE-14164
- Contrary to what Wine thought, this works even if the cabinet file does not have a ".cab" extension.
- Instead of polluting the directory where the cabinet file exists with all the files extracted from it,
we only extract the needed file to a temporary folder (thus being sure it does not overwrite any other
existing file with the same name), and then we move the extracted file to its final destination with rename.
- Translated some English strings and fixed a small grammar typo.
Addendum:
- Header file ported from ReactOS coding style added
(even though it kind of bloats the translation file, it does actually help for
future translators looking at changes committed by previous translators).
- Replaced "..." with "…" (single character) <sanchaez>
[FREELDR] Update ACPI (Extended Attributes) support in PcMemGetBiosMemoryMap()
CORE-13332
* PcMemGetBiosMemoryMap(): Add Extended Attributes set and check for entry validity and default handling of unexpected case.
* pcbios.h: Rename superceded BIOS_MEMORY_MAP.Reserved. Adapt existing code to new ACPI 6.2-A definitions.
* pcbios.h: Update BIOS_MEMORY_TYPE and BIOS_MEMORY_MAP to ACPI 6.2-A from 1.0+.
* PcMemGetBiosMemoryMap(): Misc fixes, no functional changes.
*Create PcMemCheckUsableMemorySize(), to split unrelated code out.
*Fix a copypasta in 2 output strings from ba9a1c3abb.
*Improve output readability of TRACE("ECX ...", ...).
*Move a TRACE("\n").
*Improve code style a bit.
Yes, it checks the input handle and doesn't fail if an invalid level is given, because someone may still send a Command.
This also fixes CORE-12794. Thanks for reporting!
leaves volumes in half-(dis)mounted state and thus they get remounted while not
completely dismounted.
This can lead to major breakage and data corruption.
This requires deeper fixes (let's just drop that driver!).
CORE-14124
CORE-14126
CORE-14133
So, bring back the infamous IopParseDevice() hack. Dismounting is to be fixed in FastFAT.
Even though it might not be the last remaining bug...
CORE-14124
CORE-14126
CORE-14133
In spite of what was implemented in our NT DDK sample, this is a legit operation.
This may have been turned legit starting NT5 (reminder, our implementation is
NT4 based...). So, in this situation, just return the information about the whole
disk (and not a random size) and also, mark everything with default values.
See disk_new for an example of how it works in NT5+.
CORE-14124
This is needed because ntdlltest.manifest has a dependency
on gdiplus and won't load because we don't have it in winsxs.
Eventually this must be removed.
* Remove the hacky find_actctx_dll form ldr that was taken from wine. LdrpResolveDllName and LdrpCheckForLoadedDll should get a parameter that is already redirected.
* Use RtlDosApplyFileIsolationRedirection_Ustr in LdrpLoadImportModule and LdrpUpdateLoadCount3 to redirerect the input parameter.
We applied this patch also last second in 0.4.7rls, now in master.
IntTrackPopupMenuEx() caused the BSOD when called while window is in the process of being destroyed.
This is no official system DLL and WineD3D is its only user. But latest WineD3D prefers the libtxc_dxtn codebase instead of the alternative library we used.
This is also what Mesa uses and Fedora ships now that the patents have expired, so we should stick to the same.
I'm importing the libtxc_dxtn-1.0.1 codebase from https://people.freedesktop.org/~cbrill/libtxc_dxtn/
It compiles warning-free in our tree without modifying a single line!
I'm merging this Pull Request based on the positive response in CORE-12759.
I lack a proper test case myself, but my version at least shouldn't make things worse. In my opinion, the previous one was broken anyway due to using stdcall imports when WineD3D called cdecl ones.
This also removes the NSWPAT option from our buildsystem as dxtn was its latest user.
For the record, the only place it was remaining was 1st stage, notably because
FSD (FastAT <3) was missing a few features. As this features weren't triggered
in 3rd stage (unless forced), it was not needed there any longer.
Now that they are implemented, and seem working out of the box, this hack might
not be longer anymore.
This is my ... ?! pfff attempt at killing it. It might not be the last, so just
disabling the hack for now. If there are no regressions (who can really believe that?)
we'll be clear for dropping it once for all.
it only perform a full check if dirty bit is set (which should never happen *cough*).
This is needed because otherwise, ReactOS installation takes forever when IopParseDevice is gone
due to slow ChkDsk.
This avoids issues when these partitions are formatted to FAT for setup.
For now, this commit doesn't change anything, but once IopParseDevice hack
gets removed, this will make a difference!
CORE-6305
It fixes assertion failure in vfatDestroyFCB() where we would have release parent before child.
This is still not perfect, but less bug prone...
With this commits (and ENABLE_SWAPOUT defined), ReactOS seems to unmount FAT volumes quite nice! :-)
(Tried with fsutil volume dismount X:)
* Fix Simplified Chinese Translation of EXPLORER.
* Fix Simplified Chinese Translation of MSCONFIG_NEW.
* Fix Simplified Chinese Translation of MSPAINT.
* Fix Simplified Chinese Translation of SHELL32.
* Fix Simplified Chinese Translation of SYSDM.
* Fix Simplified Chinese Translation of SETUP/REACTOS.
It's critical not to perform any write on a volume without an exclusive lock on it.
That lock is only acquired if ChkDsk is started in RW mode.
Also added an assert in write routine, to make sure that when we're about to perform
a write operation, we're really allowed to do so.
This will avoid volume corruptions when a simple "chkdsk" is issued from cmd.
To put it simple: check will really check now, and won't attempt any repair.
CORE-14119
floppy drives in ReactOS and mount images on them.
Only the cmd got imported. The GUI interface may come later on.
Note that, as for vcdrom, the driver is left disabled and you need to explicitely
start it through vfd command line interface.
CORE-14090
- The aliases ("macros") definition files have the same format as what
is output by DosKey when one displays the list of defined aliases with
the /macros: command-line switch. It should then, in particular,
understand the format where executable names are also specified:
[exename1]
alias1=command1
...
[exename2]
aliasN=commandN
...
We now behave similarly as Windows' version of this utility in that
regard.
- Use ARRAYSIZE() when needed.
- Take the opportunity to add a license header to the source file.
- Move Accessibility, Entertainment and System Tools folders into Accessories.
- Move mstsc.exe shortcut into new Accessories/Communications folder.
- Rename Remote desktop to Remote Desktop Connection.
- Rename Regedit to Registry Editor.
- Plus: Sort touched lines in the list.
* Fix the Simplified Translation of Rapps
* Fix the Simplified Translation of msconfig_new
* Fix the Simplified Translation of charmap_new
* Fix the Simplified Translation of winmine
* Fix the Simplified Translation of kbswitch
* Fix Simplified Chinese Translation of logoff
* Fix Simplified Chinese Translation of eventvwr
* Fix Simplified Chinese Translation of sysdm
* Update Simplified Chinese Translation for dxdiag
* Pass the correct handle to _RpcGetPrinter in GetPrinterW.
* Pass an empty string instead of a NULL pointer as wszComputerName to the GetPrinterLevel* functions, because this variable is later used as source for StringCbCopyExW.
* Don't check for GetLastError() == ERROR_SUCCESS in tests. Windows apparently only sets the last error in case of failure.
The Printing code should probably be changed similarly in a future commit.
Should fix CORE-14072
AddNewMftEntry() - Zero the (larger) buffer for the bitmap prior to reading the bitmap.
BrowseIndexEntries() - Check sub-nodes before checking an index entry. Read and use the index bitmap when checking sub-nodes.
+BrowseSubNodeIndexEntries() - Called for sub-nodes of an index when browsing index entries.
+NtfsDumpData() - Diagnostic function which allows for visualizing a series of bytes.
svn path=/branches/GSoC_2016/NTFS/; revision=75817
+AddBitmap() - adds a $BITMAP attribute to a file record.
+AddIndexAllocation() - adds an $INDEX_ALLOCATION attribute to a file record.
+CountBTreeKeys() - Counts the number of linked B-Tree keys.
CreateIndexBufferFromBTreeNode() - Set INDEX_NODE_LARGE if the node has sub-nodes.
CreateIndexRootFromBTree() - Simplify the usage and math of MaxIndexSize; make it only account for the cumulative size of the index entries.
+DemoteBTreeRoot() - Replaces the contents of an index root with a dummy key, and puts those contents in a new node, which is made a child of the dummy key. This is done when an index root grows too large.
+GetIndexEntryVCN() - Retrieves the VCN from an index entry.
NtfsAddFilenameToDirectory() - Fix math for MaxIndexRootSize.
NtfsInsertKey() - Add support for splitting a B-Tree node. Don't check size of index root (that will be handled later).
+SplitBTreeNode() - Called when a B-Tree node grows too large.
UpdateIndexAllocation() - Create an $I30 index allocation attribute and bitmap attribute if needed.
UpdateIndexNode() - Update children before updating the current node. Store VCN of child nodes in the index entries of their respective keys.
svn path=/branches/GSoC_2016/NTFS/; revision=75707
IncreaseMftSize() - Add some fixes. Write blank records to newly-allocated mft entries, and update $MFTMirr when finished; these changes are needed for chkdsk. Increase size by 64 records instead of 8.
+UpdateMftMirror() - Backs up the first ~4 master file table entries to the $MFTMirr file.
svn path=/branches/GSoC_2016/NTFS/; revision=75694
+AddIndexRoot() - Creates an $INDEX_ROOT attribute and adds it to a file record.
AddNewMftEntry() - Make sure the buffer used by RtlInitializeBitmap() is ULONG-aligned, and a ULONG-multiple in size, per MSDN.
AllocateIndexNode() - Calculate BytesNeeded correctly. Read $BITMAP attribute before increasing its length, in anticipation of a future commit that will check for a free bit before assigning a new index record to the end of the allocation. Use appropriate Set*AttributeDataLength() function, as $BITMAP can be resident or non-resident.
B_TREE_FILENAME_NODE - Give two members more accurate names: change "ExistsOnDisk" member to "HasValidVCN" and rename "NodeNumber" member "VCN."
+CreateEmptyBTree() - Creates a B-Tree to represent an empty directory (for AddIndexRoot).
+NtfsCreateEmptyFileRecord() - Creates an empty file record in memory, with no attributes.
CreateIndexRootFromBTree() - Fix TotalSizeOfEntries calculation.
+NtfsCreateDirectory() - Creates a file record for an empty directory and adds it to the mft.
svn path=/branches/GSoC_2016/NTFS/; revision=75692
-Don't add a preceding backslash when creating files on root.
-Use NTFS_FILE_NAME_POSIX name type if CaseSensitive option is specified.
-Don't try to create a file when a folder is requested (folder creation is still TODO).
svn path=/branches/GSoC_2016/NTFS/; revision=75671
+AllocateIndexNode() - Allocates a new index record in an index allocation.
+CreateDummyKey() - Creates the final B_TREE_KEY for a B_TREE_FILENAME_NODE. Also creates the associated index entry.
GetSizeOfIndexEntries() - Sums the size of each index entry in every key in a B-Tree node.
+SetIndexEntryVCN() - Sets the VCN of a given IndexEntry.
NtfsInsertKey() - Handle instance when the index root grows too large. If it does, add its contents to a new sub-node, and replace contents with a dummy-key whose child is the new node.
UpdateIndexAllocation() - Update index entry if a key has just been assigned a child allocation.
UpdateIndexNode() - Make sure the node exists on disk, and allocate an index record for it if it doesn't.
svn path=/branches/GSoC_2016/NTFS/; revision=75557
AddRun() - Allow for resizing the size of the data runs when the attribute isn't the last in the file record. Fix some comments.
CreateIndexBufferFromBTreeNode(), CreateIndexRootFromBTree - Fix math of IndexSize when checking if the index buffer is too large.
InternalSetResidentAttributeLength() - Allow changing the length of an attribute in the middle of a file record. Adjust the position of every attribute after the one being resized.
+MoveAttributes() - Moves a block of attributes to a new location in the file Record.
PrintAllVCNs() - Add consideration for an index allocation with a size of 0.
WriteAttribute() - Add optional parameter for a pointer to the file record being written to. If passed a file record, WriteAttribute() will skip reading the file record from disk, and will update the file record in memory before returning. This helps callers that use the file record after writing an attribute to stay in-sync with what's on disk.
svn path=/branches/GSoC_2016/NTFS/; revision=75554
-PrepareAttributeContext() - update to allocate memory for pRecord. Don't assume allocations are succeeding.
-ReleaseAttributeContext() - update to free memory for pRecord.
-InternalSetResidentAttributeLength() - Increase size of AttrContext->pRecord as needed. Update to return an NTSTATUS.
-SetResidentAttributeDataLength() - Fix bug that could occur when migrating resident attributes to non-resident if AttrContext->pRecord is too small for the new attribute.
-AddRun() - Fix a bug by reallocating AttrContext->pRecord if the record needs to be enlarged.
svn path=/branches/GSoC_2016/NTFS/; revision=75493
-CreateBTreeFromIndex() - Fix memory allocation; allocate sizeof(B_TREE_KEY) bytes for the key, not the size of the pointer. Add support for child nodes. Update parameter list.
-CreateIndexRootFromBTree() - Update Header->Flags if any of index root's keys have sub-nodes.
+CreateIndexBufferFromBTreeNode() - Converts a B-Tree node to an index buffer to be written to the index allocation.
+UpdateIndexAllocation() - Updates all of the stale nodes in an index allocation based on a PB_TREE.
+UpdateIndexNode() - Writes a B-Tree node into the index allocation.
+CreateBTreeKeyFromFilename() - Creates a PB_TREE_KEY based on a FILENAME_ATTRIBUTE.
-DestroyBTreeKey() - Destroy a child node, if present.
-DumpBTreeKey() - Dump a child node, if present.
-DumpBTreeNode() - Include Node->KeyCount in output.
-NtfsAddFilenameToDirectory() - Call UpdateIndexAllocation() prior to CreateIndexRootFromBTree().
-Update B_TREE_FILENAME_NODE with members to keep track of its existence on disk.
svn path=/branches/GSoC_2016/NTFS/; revision=75428
+PrintAllVCNs() - Diagnostic function which prints VCN of every node in an index allocation.
+GetAllocationOffsetFromVCN() - Calculates location of an index buffer from the node's VCN.
+GetInfoClassName() - Gets a string representation of an info class enumeration, to speed up development of unimplemented classes.
-NtfsSetInformation() & NtfsQueryInformation() - Use GetInfoClassName to report unhandled information classes.
-CompareTreeKeys() - Add a comment and clarify some comments.
-NtfsAddFilenameToDirectory() - Don't try to update the size of Index Root on disk if the attribute length hasn't changed.
svn path=/branches/GSoC_2016/NTFS/; revision=75424
-Add define for indexed flag for resident attributes, RA_INDEXED.
-CreateBTreeFromIndex() - Don't try to read index entries beyond attribute length.
-Move NtfsAddFileNameToDirectory() from dirctl.c to mft.c (no changes to function).
-SetResidentAttributeDataLength() - Don't try to free AttribData if it hasn't been allocated. Cast IndexRecord to PUCHAR for WriteAttribute(), and cast BitmapData to PCHAR for ReadAttribute() (make gcc happy).
-Replace magic number 16 with a define, NTFS_FILE_FIRST_USER_FILE.
svn path=/branches/GSoC_2016/NTFS/; revision=75371
-Add ATTR_RECORD_ALIGNMENT define to replace magic value of 8 when we need to adjust an attribute's beginning or length.
-Don't use int's.
-GetPackedByteCount() - Remove unused "bytes" variable.
svn path=/branches/GSoC_2016/NTFS/; revision=75288
FindAttribute() - Use * sizeof(WCHAR) in place bit-shifting wizardry.
IncreaseMftSize() - Remove two useless casts. Return error code if WriteAttribute() fails.
SetNonResidentAttributeDataLength() - Assert that AttrContext parameter describes a NonResident attribute. When migrating a resident attribute to non-resident, don't leak memory if FsRtlInitializeLargeMcb() throws an exception, and don't mark the attribute as non-resident until after the Mcb is initialized.
InternalSetResidentAttributeLength() and SetResidentAttributeLength() - Assert that AttrContext parameter describes a resident attribute.
AddNewMftEntry() - Simplify math of backing up and restoring reserved bits by changing BitmapData to PUCHAR type. Don't attempt to modify a volume with more than 2^32 clusters.
AddFixupArray() - Use ULONG in place of int.
svn path=/branches/GSoC_2016/NTFS/; revision=75284
NtfsSetEndOfFile() - Make fileNameAttribute and filename variables uppercase. Don't leak FileRecord if we can't truncate the file. Don't leak memory if there's no FileName attribute.
svn path=/branches/GSoC_2016/NTFS/; revision=75283
-NtfsOpenFile() - Replace an ExFreePool() with ExFreePoolWithTag().
-NtfsCreateFile() - Fix broken cast with BooleanFlagOn() macro.
-NtfsAddFilenameToDirectory() - Remove an extra cast. Return an error if we fail to allocate I30IndexRoot.
-NtfsGetNextPathElement(), NtfsWSubString(), NtfsGetFCBForFile() - Use PCWSTR in place of const PWCHAR or PWCHAR where it makes sense.
svn path=/branches/GSoC_2016/NTFS/; revision=75281
-CompareTreeKeys() - Assert that the first key isn't the dummy key.
-CreateIndexRootFromBTree() - Assert that CurrentKey->IndexEntry->Length isn't 0.
-DumpBTreeKey() - Use sizeof(WCHAR) in place of magic 2.
-NtfsInsertKey() - Check for allocation failure of NewKey.
svn path=/branches/GSoC_2016/NTFS/; revision=75280
-AddFileName() - Don't use dedicated buffer for FileNameNoPath, just point into FileObject->FileName's buffer. Don't use RtlCopyUnicodeString(), just update pointers.
-AddRun() - Fix an SEH2_YIELD with missing return statement.
-FreeClusters() - Use ULONG_MAX in place of 0xffffffff.
-NtfsDumpIndexRootAttribute() - Use consistent UpperCase naming convention. Use BooleanFlagOn() macro where appropriate. Replace magic 8 with sizeof(ULONGLONG).
-GetFileNameAttributeLength() - Add documentation.
svn path=/branches/GSoC_2016/NTFS/; revision=75279
NtfsAddFilenameToDirectory() - Add CaseSensitive parameter. Update to use new B-Tree code: First, the index is read and converted to a B-Tree in memory. Next, a key for the new file is inserted into the tree. Finally, the tree is converted back to an index root attribute which is written to disk.
+btree.c - Includes functions related to B-Trees (AKA B*Trees).
ntfs.h - Added several structures for representing B-Trees in memory.
Known limitations: For simplicity, only trees with a depth of one are currently supported (i.e. an ordered list of filenames). Directories that have or will require an index allocation to store all their filenames are still TODO. As a consequence, the user will only be able to create about 6 files in a directory.
svn path=/branches/GSoC_2016/NTFS/; revision=75223
Improve, add, or fix some DPRINTs. In particular, ULONG's should use %lu, not %u. Also, don't be silent about filesystem corruption.
NtfsFindMftRecord() - move CaseSensitive parameter before output parameter in parameter list.
svn path=/branches/GSoC_2016/NTFS/; revision=75199
Whether or not a filename should be interpreted as case-sensitive is dependent on a flag passed to the driver when a file is created (opened); it's separate from the namespace associated with the file being accessed.
svn path=/branches/GSoC_2016/NTFS/; revision=75178
-AddNewMftEntry() - Increase size of MFT as needed. Fix math for bitmap length. Don't assign file records to MFT indices 0x10 - 0x17; In Windows, these records aren't used unless they have to be, even though they are marked as unused in the bitmap.
+IncreaseMftSize() - Adds room for additional file records in the master file table.
-BrowseIndexEntries() - allow for the rare situation when a non-system file has an MFT index of 0x10.
svn path=/branches/GSoC_2016/NTFS/; revision=75056
-SetAttributeDataLength() has been split into two functions, SetNonResidentAttributeDataLength() and SetResidentAttributeDataLength(). This should improve code readibility and allows for resizing an attribute when there's no FileObject associated with it.
-Added "MftDataOffset" member to DEVICE_EXTENSION, which stores the offset of the Mft's $DATA attribute. (I'm starting to think it's better to add a member for offset to NTFS_ATTR_CONTEXT directly, but I'll save that level of restructuring for a future commit.)
svn path=/branches/GSoC_2016/NTFS/; revision=75055
AddFileName() - Add a parameter to receive the Mft index of the parent directory. Fix so the name of the file will be stored in the attribute, not the name of the directory.
NtfsCreateFile() - Open a file that was successfully created, instead of assuming failure.
NtfsCreateFileRecord() - Add the filename attribute of the created file to the parent directory's index.
+NtfsAddFilenameToDirectory() - Adds a $FILE_NAME attribute to a given directory index. Presently, a file can be created in an empty directory only.
AddNewMftEntry() - Add a parameter to receive the mft index where the new entry was stored.
svn path=/branches/GSoC_2016/NTFS/; revision=74970
[BOOTDATA] - Add a commented-out section to hivesys.inf which can add the required key to enable NTFS write support.
svn path=/branches/GSoC_2016/NTFS/; revision=74685
-Handle IRP_MJ_SET_INFORMATION IRP requests.
+NtfsSetEndOfFile() - Sets the end of file (file size) for a given file.
+NtfsSetInformation() - Sets the specified file information. At this point, only FileEndOfFileInformation is fully implemented. FileAllocationInformation is handled the same way and not truly implemented, but this works well enough for SetEndOfFile().
Overwriting a file in NTFS should now work in the majority of use cases.
svn path=/branches/GSoC_2016/NTFS/; revision=74675
-Modify SetAttributeDataLength() to allow a resident attribute to migrate to non-resident if the attribute grows too large to remain resident.
-Fix values returned by NtfsAllocateClusters() in case of error; return error codes, not 0.
svn path=/branches/GSoC_2016/NTFS/; revision=74524
-Add a new member to the NTFS_ATTR_CONTEXT struct, a LARGE_MCB. This allows an attribute context to describe the cluster mapping of a non-resident file while allowing that mapping to change dynamically, without the context itself needing to be resized. This fixes problems which sometimes arose from resizing files.
-Remove hacky code from NtfsWriteFile() for dealing with "stale" contexts. This fixes that issue.
-Update SetDataAttributeLength(), PrepareAttributeContext(), ReleaseAttributeContext(), FreeClusters(), and AddRun() for the new member.
-Update ReadAttribute() and WriteAttribute() to work with the changed structure. A very-soon-to-come commit will overhaul these functions so they'll operate directly on the LARGE_MCB, instead of converting to and from a packed list of data runs. (Sparse files are broken until then.)
-Rename "RunBufferOffset" to "RunBufferSize" in several places where appropriate.
-Fix, improve, and add some comments.
svn path=/branches/GSoC_2016/NTFS/; revision=74523
+AddStandardInformation(), +AddData(), +AddFileName() - Add attributes to a file record
+NtfsCreateFileRecord() - Creates a new file record and saves it to the MFT.
+AddNewMftEntry() - Adds a file record to the MFT.
NtfsCreateFile() - Modified to create a file record on a file-creation request (file creation is still unsupported; the created file needs to be added to the parent's directory index).
+SetFileRecordEnd() - Establishes a new file record size
UpdateFileRecord() - Improved documentation
InternalSetResidentAttributeLength() - Updated to use SetFileRecordEnd().
svn path=/branches/GSoC_2016/NTFS/; revision=74321
Add some fixes to attrib.c, as suggested by Pierre Schweitzer:
*ConvertLargeMCBToDataRuns() - Use MS portable type, ULONG, for variable i.
*FreeClusters(), AddRun() - Check for invalid parameter before allocating memory, and confirm the memory is allocated.
*ConvertDataRunsToLargeMCB(), AddRun() - Avoid code duplication by using ExRaiseStatus() in try block, and return accurate status via _SEH2_GetExceptionCode().
svn path=/branches/GSoC_2016/NTFS/; revision=72422
Implement AddRun(). Add support functions and documentation.
+ConvertDataRunsToLargeMCB()
+ConvertLargeMCBToDataRuns()
*SetAttributeDataLength(), *NtfsWriteFile() - Update for AddRun() implementation. Add hack to SetAttributeDataLength() to allow notepad.exe to save files until freeing clusters is implemented.
svn path=/branches/GSoC_2016/NTFS/; revision=71942
Add support functions, fix warning from previous commit.
+NtfsDumpDataRuns(), +NtfsDumpDataRunData() - Provide diagnostic output.
+GetPackedByteCount() - Used to encode data runs.
svn path=/branches/GSoC_2016/NTFS/; revision=71921
Add ability to write to resident attributes.
SetAttributeDataLength() - Check if the file is memory mapped before truncating
+InternalSetResidentAttributeLength() - Used by SetAttributeDataLength()
svn path=/branches/GSoC_2016/NTFS/; revision=71820
Allow for an existing file to be opened with FILE_OVERWRITE, FILE_OVERWRITE_IF, or FILE_SUPERSEDE dispositions, and truncate that file. This allows for a file to be opened and saved in Notepad.exe [provided that file is non-resident and its allocation size doesn't need to change].
svn path=/branches/GSoC_2016/NTFS/; revision=71680
Update a file's size in the relevant $FILE_NAME attribute of the index entry in the parent directory.
+UpdateFileNameRecord() - Searches a parent directory for the proper index entry, then updates the file sizes in that entry.
+UpdateIndexEntryFileNameSize() - Recursively searches directory index and applies the size update.
svn path=/branches/GSoC_2016/NTFS/; revision=71664
Fix Up AddFixupArray - It needs to accept a PNTFS_RECORD_HEADER for parameter 2, not a PFILE_RECORD_HEADER.
svn path=/branches/GSoC_2016/NTFS/; revision=71662
When writing to a file, increase the file size if trying to write past the end.
*FindAttribute() has been given an optional pointer to a ULONG that will receive the offset of the found attribute from the beginning of the record. This is to allow for found attributes to be written back into their file records.
+SetAttributeDataLength()
+UpdateFileRecord() - Updates a file record in the master file table at a given index.
+AddFixupArray() - Prepares a file record or directory index for writing to the disk.
svn path=/branches/GSoC_2016/NTFS/; revision=71660
Add fixes to WriteAttribute():
-Remove erroneous check for end of run (before writing to the last run returned)
-Properly dereference RealLengthWritten pointer, as reported in CR-90
svn path=/branches/GSoC_2016/NTFS/; revision=71616
This is likely due to a copy paste error where long name was copied twice and short never.
Fun fact: this was not affecting FATX volumes
Fun fact2: this was defeating a buffer overflow check and thus was allowing buffer overflow!
CORE-14088
This makes use of previous commit vfatlib improvements that allow
caller to stop chkdsk if locking fail, which will happen if volume is in use.
That way, ReactOS users won't be able any longer to f*** up their C:\ volume
by attempting to chkdsk -f it!
CORE-14087
For instance, when repair is required, we can ask the caller whether we should
continue or not in case locking failed.
Also, introduced a hack for 1st stage where IopParseDevice() hack is in usage
so that broken NTSTATUS is diverted to appropriate status.
That way, usetup will properly continue even if locking failed (due to its
callback stub!)
CORE-14087
- We use this as a chance to also update the CMake file in terms of consistency.
- The clang-cl build applies the /Os flag on all the source files that are now using the PCH too.
This may get more applications to work which previously called into the WINE-generated stubs (that throw an exception).
It also improves debug logs when the "winspool" debug channel is enabled. With such detailed debug logs, we may get an idea which winspool APIs need to be implemented sooner than others.
This commit fixes weird behavior in our FastFAT implementation. It was mixing two classes:
FileNameInformation and FileNamesInformation. It was handling FileNameInformation like
FileNamesInformation and was filling buffer with FILE_NAMES_INFORMATION structure instead
of FILE_NAME_INFORMATION structure (how many things did that break?!).
Also, it wasn't implementing the FileNamesInformation class at all. This is required by
ntdll_winetest:directory which doesn't expect it to fail and thus, attempts to read
never filled in memory.
This commit fixes the winetest crash, and may fix other weird FS behavior.
CORE-13367
The current implementation was broken and some dead code was never called; always failing before.
This fix revives this dead code!
Extra fix: avoid derefencing potential null-ptr.
And also, as bonus, comment the function so that logic can be easily understood.
CORE-14062
- Add Simplified Chinese translation for EVENTCREATE.
- Update translations for 'dxdiag', 'fontview', 'magnify', 'mmc', 'rapps' and 'wordpad'.
Patch by Li Keqing.
Its purpose is to dump the non paged consumption, tag by tag,
to allow tracking potential faulting driver in case ReactOS starts lacking memory.
This will look like what !poolused outputs, even though it doesn't deal with paged pool.
Thanks to Thomas for his kind review and improvement suggestions.
CORE-14048
- Avoid a potential race whereby the current service selection can change before the propsheet thread starts up
- Cleanup the depends data, it doesn't need to be passed around the propsheet
By providing /u or /j switch on mount, the virtual CD-ROM class driver
will break the associated signatures on read so that FSDs cannot recognize them any longer.
[SERVMAN]
- Make the property sheets modeless so users can open multiple services at the same time
- Untested in ros. In fact we have no code or tests cases to check that modeless property sheets work, so please raise a bug if you find any issues with the app.
- Dedicated to reactosfanboy
It allows you managing virtual drives, mounting images, ejecting images, and so on.
It will load the driver when required.
It's less features complete that the MS GUI implementation.
I'll implement them later on, it can be quickly done.
What's missing: UDF/Joliet suppr and persistent mounts
It was provided by MS as a separate package to download for Windows (up to 7).
That class driver allows creating virtual drives on which we can later mount ISOs images.
It's rather basic, but does the job.
To use it, you can use the GUI app from Microsoft (Virtual CD-ROM Control Panel)
or the vcdcli in CLI. We're compatible :-).
Note that it's not loaded at boot, you need to manually start it, to lower memory footprint.
Both applications will handle this for you.
The command should delete files in the specified directory and all of its
sub-directories, using any file pattern specified.
For example, the command:
del /S .\my_directory
should delete all the files inside my_directory and its sub-directories,
and as such should also detect that "my_directory" is indeed a directory,
while doing:
del /S .\my_file
should of course detect that "my_file" is indeed a file pattern, and thus,
delete all "my_file" files from the current directory and its sub-directories.
The command:
del /S some_directory\file_pattern
should delete "file_pattern" files from some_directory and its sub-directories.
Unified the first badge row to use the same colors as reactos.org
Moved build related badges to 'Build' section
Fixed all links to use https and to be semantically correct
Replaced 'Tweet' badge with 'Follow', 'Flattr' with more general 'donate'
Added a sentence about donation
Centered second badge row by using HTML code
Moved Coverity scan to the top, Flattr and Tweet - to the bottom
Linked "Download" and "Release" to reactos.org
Be respectful toward other members of the community. Don't initiate or engage in discussions that are designed to insult, inflame, attack, or incite hate/discrimination/bullying against others. Don't feed the trolls. If we detect any kind of behaviour we consider unacceptable, we reserve the right to ban, block, or remove the involved members from any or all of our communities.
# See Also
- [Rules for managing Pull Requests](PULL_REQUEST_MANAGEMENT.md)
@@ -49,7 +49,7 @@ Your contribution can be of numerous forms. We currently accept two ways to cont
### Pull Requests
Since our [migration to GitHub] we gladly accept __[Pull Requests]__. Pull requests let you tell others about changes you have pushed to a repository on GitHub. Once a pull request is opened, you can discuss and review the potential changes with collaborators and add follow-up commits before the changes are merged into the repository. __Pull request is a prefered way to submit your work__ - it makes reviewing and merging your contribution much easier.
Since our [migration to GitHub] we gladly accept __[Pull Requests]__. Pull requests let you tell others about changes you have pushed to a repository on GitHub. Once a pull request is opened, you can discuss and review the potential changes with collaborators and add follow-up commits before the changes are merged into the repository. __Pull request is a preferred way to submit your work__ - it makes reviewing and merging your contribution much easier.
### Patches
@@ -57,6 +57,10 @@ A __[patch]__ is a set of changes to existing source code. The changes in a patc
See [Submitting Patches] for details.
### Commit style
Our commit style is defined in a __[commit template]__. Use it as a reference or turn it on using `git config commit.template .gitmessage`. This will set this template as an initial commit message for the new commits in your local repository.
### Rules and Recommendations
- *Use your __real name__ and __real email__.* We do not accept anonymous contributions!
@@ -67,6 +71,8 @@ See [Submitting Patches] for details.
- *Respect our __[Coding Style]__ and __[Programming Guidelines]__.*
- *Do not be afraid to ask questions.* Ask our developers on JIRA or [IRC] channel.
To amend your commit with your name and e-mail (in any case you've forgot to set your name/e-mail) please take a look at this [guide](https://reactos.org/wiki/ReactOS_Git_For_Dummies#Amending_your_commit_with_name.2FE-mail). To set your name/e-mail globally for future commits that you push, [read this](https://reactos.org/wiki/ReactOS_Git_For_Dummies#Assign_commits_with_your_name_.26_E-mail_automatically).
## Where To Start?
Finding a good project to start with can be a challenge, because when starting out you are (usually) not aware of all the possibilities. To help you find a project, here are some ideas to try:
@@ -96,3 +102,8 @@ Finding a good project to start with can be a challenge, because when starting o
For the sake of trying to maintain an acceptable number of open but idle PRs, the following rules should be considered:
- If a PR has at least one approval, it can be merged after 1 week of waiting for additional comments.
- If the change has at least 3 approvals or you consider it trivial enough, it may be merged right away.
- If a PR stays in "changes requested" for too long, and there is no indication from the author that they are working on it, it shall be closed.
- Rule of thumb: 2 weeks for a small PR. Can be longer if the PR is large.
- The PR can be reopened at any point, if you have additional comments, or new changes have been done.
- If you require a review from a particular person, assign the PR to that person. Don't just rely on the "review requested" feature of GitHub.
- Remember that PR labels exist. You can assign an appropriate label to a pull request to designate it's scope, grab additional attention or just for extra navigation possibilities.
- Don't feel obliged to comment everything you see, just for the sake of commenting. Be it on JIRA, GitHub, or even on IRC.
In addition, in order to avoid coming off as rude to helpful contributors, please refrain from:
- Asking the contributor to do unrelated work
- Closing without providing a reason
- Merging with the intention to rewrite that code soon after
@@ -21,14 +43,18 @@ ReactOS™ is an Open Source effort to develop a quality operating system that i
The ReactOS project, although currently focused on Windows Server 2003 compatibility, is always keeping an eye toward compatibility with Windows Vista and future Windows NT releases.
The code of ReactOS is licensed under [GNU GPL 2.0+](https://spdx.org/licenses/GPL-2.0+.html).
The code of ReactOS is licensed under [GNU GPL 2.0](https://github.com/reactos/reactos/blob/master/COPYING).
## Building [![rosbewin.badge]][rosbewin.link] [![rosbeunix.badge]][rosbeunix.link]
***ReactOS is currently an Alpha quality operating system. This means that ReactOS is under heavy development, things may not work well and it can corrupt the data present on your hard disk. It is recommended to test ReactOS on a virtual machine or on a computer with no sensitive or critical data!***
To build the system it is strongly advised to use the _ReactOS Build Environment (RosBE)._
Up-to-date versions for Windows and for Unix/GNU-Linux are available from our download page at: http://www.reactos.org/wiki/Build_Environment.
Up-to-date versions for Windows and for Unix/GNU-Linux are available from our download page at: ["Build Environment"](http://www.reactos.org/wiki/Build_Environment).
Alternatively one can use Microsoft Visual C++ (MSVC) version 2010+. Building with MSVC is covered here: https://www.reactos.org/wiki/Building_with_MSVC.
Alternatively one can use Microsoft Visual C++ (MSVC) version 2010+. Building with MSVC is covered here: ["Visual Studio or Microsoft Visual C++"](https://www.reactos.org/wiki/CMake#Visual_Studio_or_Microsoft_Visual_C.2B.2B).
### Binaries
@@ -41,28 +67,36 @@ build directory. This will create a CD image with a filename `bootcd.iso`.
See ["Building ReactOS"](http://www.reactos.org/wiki/Building_ReactOS) for more details.
You can always download fresh binary builds of bootable images from the ["Daily builds"](https://www.reactos.org/getbuilds/) page.
## Installing
ReactOS currently can only be installed on a machine that has a FAT16 or FAT32 partition as the active (bootable) partition.
The partition on which ReactOS is to be installed (which may or may not be the bootable partition) must also be formatted as FAT16 or FAT32.
By default, ReactOS currently can only be installed on a machine that has a FAT16 or FAT32 partition as the active (bootable) partition.
The partition on which ReactOS is to be installed (which may or may not be the bootable partition) must also be formatted as FAT16 or FAT32.
ReactOS Setup can format the partitions if needed.
Starting 0.4.10, ReactOS can be installed using the BtrFS file system. But
consider this as an experimental feature and thus regressions not triggered on
FAT setup may be observed.
To install ReactOS from the bootable CD distribution, extract the archive contents. Then burn the CD image, boot from it, and follow the instructions.
See ["Installing ReactOS"](https://www.reactos.org/wiki/Installing_ReactOS) Wiki page or [INSTALL](INSTALL) for more details.
## Testing
If you discover a bug in ReactOS search on JIRA first - it might be reported already. If not report the bug providing logs and as many information as possible.
If you discover a bug in ReactOS search on JIRA first - it might be reported already. If not report the bug providing logs and as much information as possible.
See ["File Bugs"](https://www.reactos.org/wiki/File_Bugs) for a guide.
__NOTE:__ The bug tracker is _not_ for discussions. Please use `#reactos` Freenode IRC channel or our [forum](https://reactos.org/forum).
## Contributing
## Contributing ![prwelcome.badge]
We are always looking for developers! Check [how to contribute](CONTRIBUTING.md) if you are willing to participate.
You can also support ReactOS by [donating](https://reactos.org/donating)! We rely on our backers to maintain our servers and accelerate development by [hiring full-time devs](https://reactos.org/node/785).
## More information
ReactOS is a Free and Open Source operating system based on the Windows architecture,
@@ -72,43 +106,30 @@ It is not another wrapper built on Linux, like WINE. It does not attempt or plan
ReactOS is also not "yet another OS". It does not attempt to be a third player like any other alternative OS out there. People are not meant to uninstall Linux and use ReactOS instead; ReactOS is a replacement for Windows users who want a Windows replacement that behaves just like Windows.
More information is available at: https://www.reactos.org.
More information is available at: [reactos.org](https://www.reactos.org).
Also see the [media/doc](/media/doc/) subdirectory for some sparse notes.
## Who is responsible
Active devs are listed as members of [GitHub organization](https://github.com/orgs/reactos/people).
Active devs are listed as members of [GitHub organization](https://github.com/orgs/reactos/people).
See also the [CREDITS](CREDITS) file for others.
## Code mirrors
The main development is done on [GitHub](https://github.com/reactos/reactos). We have an [alternative mirror](https://git.reactos.org/) in case GitHub is down.
The main development is done on [GitHub](https://github.com/reactos/reactos). We have an [alternative mirror](https://git.reactos.org/?p=reactos.git) in case GitHub is down.
There is also an obsolete [SVN archive repository](https://svn.reactos.org/svn/reactos?view=revision) that is kept for historical purposes.
There is also an obsolete [SVN archive repository](https://svn.reactos.org/reactos/) that is kept for historical purposes.
IDS_LICENSE "This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.\r\n\r\nThis program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.\r\n\r\nYou should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA."
LTEXT "Hartă de caractere v0.1\nDrept de autor (C) 2007 Ged Murphy (gedmurphy@reactos.org)\nTraducere de Ștefan Fulea (stefan.fulea@mail.md)", IDC_STATIC, 48, 7, 210, 36
LTEXT "Hartă de caractere v0.1\nDrept de autor (C) 2007 Ged Murphy (gedmurphy@reactos.org)\nTraducere de Ștefan Fulea (stefan.fulea@mail.com)", IDC_STATIC, 48, 7, 210, 36
ERROR_UNSUPPORTED_FORMAT "Pano görüntülenemeyen bir biçimdeki veri içeriyor."
ERROR_INVALID_FILE_FORMAT "Seçili kütük geçerli bir pano kütüğü değil."
ERROR_INVALID_FILE_FORMAT "Seçili dosya geçerli bir pano dosyası değil."
END
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
