Disable enqueuing processes into the SessionGlobal->ProcessList, since it's broken and causes non paged pool corruption. The reason is that sometimes the process is not removed, so a dead allocation stays in the list, causing havoc, when the next or previous process in that list is removed.
svn path=/trunk/; revision=58487
- Implement "!pool" kdbg extension that works like in WinDbg
- Implement ExpCheckPoolAllocation to check a single allocation for sanity
- Rename MEMORY_ARE::PageOpCount (which is unused) to Magic and set it to 'MAre'
- Implement MiRosCheckMemoryAreas, that checks all memory areas for integrity and PspCheckProcessList that checks the process list. The code is not used anywhere, since it can cause a major performance impact, but it proved useful to detect non paged pool corruptions. So I'll commit it in the hope that it will be useful.
svn path=/trunk/; revision=58485
- Implement SamIConnect.
- Add a trusted caller flag to the database object type and inherit it to opened or created sub objects.
- Restrict access to SamrQueryInformationUser.UserInternal1Information for non-trusted callers.
svn path=/trunk/; revision=58472
In LsapCreateDatabaseObjects pass the pointer AuditEventsInfo to LsapSetObjectAttribute instead a pointer to the pointer. Fixes 1 advapi32::lsa winetest when compiled with the new BE.
svn path=/trunk/; revision=58461
- Implement LSA dispatch routines LsapAllocateClientBuffer, LsapFreeClientBuffer, LsapCopyToClientBuffer and LsapCopyFromClientBuffer.
- LsapCallAuthenticationPackage and LsapLogonUser: Copy data from the callers address space into a local buffer and pass it to the authentication package.
svn path=/trunk/; revision=58457
- Move LsapLookupAuthenticationPackage, LsapCallAuthenticationPackage and LsapLogonUser from authport.c to authpackage.c.
- Make LsapCallAuthenticationPackage and LsapLogonUser call the matching authentication package functions.
svn path=/trunk/; revision=58451
* Fix some format specifiers (part 2 of r58442). Thanks Amine :)
* Remane two "global" static variables to avoid name collision warnings.
svn path=/trunk/; revision=58443
* Add "/nologo" to the MSVC resource compilation rule. Brought to you by Yu Pan (binsys {at} 163 dot com).
CORE-6997 #resolve #comment Committed in r58438.
svn path=/trunk/; revision=58438
- Fix a bug in EngLoadModuleEx
- Fix a bug in co_IntLoadSysMenuTemplate
- Fix / improve a number of annotations
- Improve code in NtGdiExtGetObjectW
- Check return value of ZwAllocateVirtualMemory and handle error in GdiPoolAllocate
- Fix possible memory leaks in NtGdiPolyDraw
- Check for NtGdiExtCreatePen == NULL instead of passing it to ProbeForRead in NtGdiExtCreatePen
- Simplify code in NtGdiGetTextMetricsW
- Fix a number of format specifiers
svn path=/trunk/; revision=58434
- Get all procedure addresses of the loaded authentication packages and keep them in the authentication package entry.
- Implement parts of the lda dispatch table (allocate heap and free heap) and pass the table to LsaApInitializePackage call.
- Implement authentication package lookup by name.
[MSV1_0]
- Store the dispatch table entries passed to the LsaApInitializePackage call and use them to allocate a STRING to return the package name.
svn path=/trunk/; revision=58427
Properly handle glyph positions and take target coordinate transformation into account when caching glyphs. (This is suboptimal, since we might cache duplicates of the same glyphs, but it works). Fixes CORE-4657.
svn path=/trunk/; revision=58418
Remove LsaRegisterPolicyChangeNotification and LsaUnregisterPolicyChangeNotification because they are not exported by advapi32.dll. These functions are exported by secur32.dll instead.
svn path=/trunk/; revision=58413
- Remove CsrSetLogonNotifyWindow
- Use a reactos only TWOPARAM_ROUTINE_EXITREACTOS call in order to start the shut down process in winlogon
- Use a NOPARAM_ROUTINE_ISCONSOLEMODE call in DtbgIsDesktopVisible to fix detecting console mode in gui-on-demand mode
svn path=/trunk/; revision=58411
- Fix the "grey panel" bug when doing File --> New.
- Add the possibility, when doing File --> New, to save the drawing the user did before actually reset it, and take into account whether he/she cancels the operation in the middle.
Patch by Marco Vaiano, slightly modified by me.
CORE-5642 #resolve #comment Committed in revision r58408. Thanks :)
svn path=/trunk/; revision=58408
- Replace the old LogonStatus variable by a LOGON_STATE enum type.
- Declare and document all new states and their transitions.
svn path=/trunk/; revision=58391
- Make NtUserSwitchDesktop hide the previous desktop window and show the new one. This doesn't change much yet as winlogon still uses only one desktop
svn path=/trunk/; revision=58386
CORE-6977 #comment Make sure the buffer for NtQueryInformationAtom is ULONG aligned. Fixes failed atom tests on ReactOS and Windows when compiled with GCC 4.7.2. Should be sent to wine. Keeping the jira issue open.
svn path=/trunk/; revision=58378
CORE-6969 Adding a check to skip a test after failure instead of continuing with uninitialized memory. "Fixes" a crash when compiled with GCC 4.7.2
svn path=/trunk/; revision=58377
- Do not set the current thread desktop automatically the first time CreateDesktop is called. The caller is supposed to call SetThreadDesktop to specify which desktop should be used
svn path=/trunk/; revision=58376
Get rid of reactos_arc_hardware_data, which wasn't used anyway. This frees up 16KB of memory. Don't reallocate the PARTIAL_RESOURE_LISTs in FldrCreateComponentKey, instead take over the ones passed by the caller.
svn path=/trunk/; revision=58369
- Fix a bug in StartupWindowThread that had as a result to close the desktop handle we get from winlogon. The problem is the wrong usage of SetThreadDesktop as it closes the previous desktop, so keeping OldDesk is totally useless and when the thread exits it will close the handle to the current thread desktop which is Session->WinlogonDesktop from winlogon. To solve this issue we have to duplicate the handle we get from winlogon and let it get closed by win32k when the thread exits
svn path=/trunk/; revision=58367
- Fix a nasty bug in IntTrackMouseMove that prevented us from stopping the ID_EVENT_SYSTIMER_MOUSEHOVER timer.
- IntKillTimer requires a PWND as first argument but we were passing an HWND and as a result the timer was never stopped
- Fixes the remaining failed tests for TrackMouseEvent (besides queue state tests)
svn path=/trunk/; revision=58364
- Implement WlxScreenSaverNotify. I decided to return a hardcoded FALSE because RegQueryValueExW always reads "1", although there is a "0" written in the registry. Will fix it later.
svn path=/trunk/; revision=58363
- Use the "Fixedsys Excelsior 3.01-L2" font instead of "Fixedsys Excelsior 3.01" for the Fixedsys font substitution (the *-L2 version corrects a bug when displaying the letter 'A'-upper-case).
- Add the font-substitute "Lucida Console" to "DejaVu Sans Mono".
svn path=/trunk/; revision=58362
When processing forwarders and the target image was not yet loaded, try to load it (using .dll extension if no extension is given)
svn path=/trunk/; revision=58361
- Remove the SCRNSAVE.EXE value from the registry and set ScreenSaveActive to "0" as this is the setting for "None" screen saver.
- Change screen saver settings the way Windows XP does it.
svn path=/trunk/; revision=58360
- Call DispatchSAS when the screen saver is started by a WM_LOGONNOTIFY message.
- Lock the workstation after the screen saver was terminated.
svn path=/trunk/; revision=58359
Remove calls to SystemParametersInfoW(SPI_SETSCREENSAVEACTIVE,... as it does not make sense to activate and deactivate screen savers when activating or deactivating the screen saver page. The screen saver should be activated when the user chooses a screen saver and it should be deactivated when the user chooses the "None" option.
CORE-6963#resolve
svn path=/trunk/; revision=58358
- hopefully put an end to this embarassing succession by adding missing linkage to msvcrt where needed.
Than you ld, for getting symbols out of thin air
svn path=/trunk/; revision=58356
- Do not assume all zeroes in PE sections that contain uninitialized data. Such sections can have other contents as well, and MiReadPage already provides zeroes for outside-of-file data anyway
CORE-6958 #resolve
svn path=/trunk/; revision=58344
- Remove the WM_SYSMENU style from the security dialog.
- Add the unlock dialog.
- Fix the layout of the german logon dialog. Password and user name are left-aligned again.
svn path=/trunk/; revision=58328
- Fix a typo in the german lock dialog.
- Store the user name and domain in the gina context and use them in the lock dialog.
- Initialize the LOGONSERVER environment variable.
svn path=/trunk/; revision=58326
Create a context struct when the client is registered and delete it when the client is deregistered. In the mean time, pass the context to all client authentication functions. The context contains a handle to the client process and a handle to the current connection port. The process handle will later be used to retrieve and return logon data and to pass the toke handle back to the client.
svn path=/trunk/; revision=58308
- Add the LSA_CONNECTION_INFO struct to the LSA_API_MSG in order to handle the connection info in an easy way.
- Let the connection handler return OperationalMode and Status values. This obsoletes a separate register call to lsasrv.dll, thus remove the register call code entirely.
- Add dispatcher code and stubs for the remaining client calls.
- Add experimental implementation of LsaCallAuthenticationPackage and LsaLogonUser.
svn path=/trunk/; revision=58304
Make the call sequence LsaRegisterLogonProcess, LsaLookupAuthenticationPackage, LsaDeregisterLogonProcess work without failures or loss of connection. WIP.
svn path=/trunk/; revision=58303
* Rely on CMAKE_HOST_WIN32 instead of CMAKE_HOST_SYSTEM_NAME which, sometimes, isn't set for some reason. Should fix the incorrect MINGW_TOOLCHAIN_PREFIX value issue.
svn path=/trunk/; revision=58292
* Fix a corner case when the start segments overlaps the end segment in Chord(), Pie(), Arc() and ArcTo() functions.
* This should fix the Pie chart in "Disk Properties" when all the space is used.
* Brought to you by Victor Martinez Calvo.
CORE-6848 #resolve #comment Committed in r58291. Thanks !
svn path=/trunk/; revision=58291
* Account for the fact that PID_SECURITY is defined in propidl.h with a different value. Spotted by Sylvain.
CORE-6930 #resolve #comment Fixed in r58289.
svn path=/trunk/; revision=58289
- Actually fail when the maximum number of reparse attempts is reached in ObpLookupObjectName. It previously returned a pointer to whatever object was currently in charge of parsing, but without taking a reference to it.
- Fixes 'Header->HandleCount == 0' assertion failure (on \REGISTRY object) after running ntdll_winetest:reg three times (and also fixes the remaining test failure)
svn path=/trunk/; revision=58281
* Add transparency to icons in the task bar and in the tray bar. Brought to you by Carlo Bramini.
CORE-6938 #resolve #comment Committed in r58279. Grazie ;)
svn path=/trunk/; revision=58279
Fix build, but remove support for architecture specific inf sections. This means livecd cannot have those until someone implements this. I give up...
svn path=/trunk/; revision=58276
- Implement support for architecture specific INF sections both in mkhive and in usetup (For an example, check hivesys.inf at the bottom)
- Get rid of architecture specific .inf files
svn path=/trunk/; revision=58271
- Add interface function to IUSBHardwareDevice, which returns the USB controller type
- The controller type is used for debugging in the common usb library (libusb)
svn path=/trunk/; revision=58268
- Ignore the unsupported MEM_TOP_DOWN flag instead of failing NtAllocateVirtualMemory
CORE-6392 CORE-6805 #comment MEM_TOP_DOWN should be sufficiently fixed in r58266, and setup now starts. Please confirm.
svn path=/trunk/; revision=58266
- Process startup items before creating the desktop. Fix some comments, add TODOs. Based on patch by Edijs Kolesnikovičs.
CORE-6896 #resolve
svn path=/trunk/; revision=58262
SamSetInformationUser: Build the LM password and calculate the LM password hash in the UserSetPasswordInformation case.
svn path=/trunk/; revision=58259
Remove a comment becoming pointless since revision r58255.
[WIN32SS]
Remove an operation, redundant since revision r58255.
svn path=/trunk/; revision=58257
SamSetInformationUser: Add special code for the UserSetPasswordInformation class that calculates the NT hash of the password and stores it by calling SamrSetInformationUser using the UserInternal1Information class.
svn path=/trunk/; revision=58256
- Remove x64 asm stub for acos from cmake file, since we already have a generic C implementation
- Implement sqrt for amd64 in SSE, both in C and asm. While the C version would be sufficient, it's currently less portable due to the lack of mm intrinsics for GCC
- Silence a warning
svn path=/trunk/; revision=58251
- Add a DPRINT with more info to the 'Hash == TableMask' assertion
- Validate pool tags on free also for big allocations
- Validate pool tag earlier, before tracking the free operation
CORE-6929
svn path=/trunk/; revision=58246
Implement IoCancelFileOpen().
You can read: http://www.osronline.com/showThread.cfm?link=20807
Even though the proposed implementation is closer to W2K implementation than to W2K3 implementation
In W2K3, no IRP is issued with major close.
svn path=/trunk/; revision=58244
Implement IopAllocateIrpMustSucceed() which is designed to *normally* always return an IRP.
Even in low memory situations (if you wait enough).
svn path=/trunk/; revision=58243
Revert my change of loading the win32k kernel-mode subsystem driver by reading the Kmode value and use instead a hardcoded value, per request of Alex.
I'm wondering why Windows wouldn't allow you to use a personalized value there and forces you to use win32k with this name, placed in \SystemRoot\System32 directory and not in an other place.
svn path=/trunk/; revision=58241
- Uniformize the error messages and also, SmpParseCommand is SmpParseCommandLine in fact.
- Remove an unneeded cast since Flags is already ULONG.
svn path=/trunk/; revision=58238
Fix acquiring/releasing driver-loading and unloading privilege (see r57754) by using the RtlAdjustPrivilege API instead of using the standard user-mode Win32 API (thus get rid of calling LSA when calling LookupPrivilegeValueW).
svn path=/trunk/; revision=58233
- Actually put in the C++ headers what our CRT implements, instead of some random mix of libstdc++ and who knows what. This means the header contents are somewhat outdated and non-standard, but that's preferable to unusable
- Fix a forward declaration in math.h
svn path=/trunk/; revision=58228
Fix a fixme in IopParseDevice() by calling SeOpenObjectAuditAlarm().
Won't change anything right now since it calls a stub.
svn path=/trunk/; revision=58226
- Use set_cpp instead of hand-coding all the parameters
- Use comutil.h and link with comsupp unconditionally, since we now have them
- Use STLPort-compatible filebuf constructor for TCHAR iostreams
- Fix a few MSVC and GCC 4.7 errors/warnings
svn path=/trunk/; revision=58224
- Import C++ compiler support headers from mingw-w64 (with some fixes)
- Implement the required parts of the comsupp library
svn path=/trunk/; revision=58223
- Do not allow multiple "Taskbar and Start Menu Properties" windows. Based on patch by Edijs Kolesnikovičs & Grégori Macário Harbs.
CORE-6885 #resolve
svn path=/trunk/; revision=58217
- Add missing domain information types to ntsam.h and fix conflicts in ntsecapi.h and sam.idl.
- Add SamChangePasswordUser, SamChangePasswordUser2 and SamChangePasswordUser3 stubs.
svn path=/trunk/; revision=58216
* This significantly shrinks the dependency tracking data and the compiled objects size, which speeds up the whole build process (especially the incremental builds) accelerating the daily development as a result.
* No intended code changes, ~27% smaller build. Enjoy ;)
svn path=/trunk/; revision=58214
- Do not allow multiple run dialogs. Based on patch by Edijs Kolesnikovičs & Grégori Macário Harbs.
CORE-6871 #resolve
svn path=/trunk/; revision=58208
- Remove duplicate dllimport attributes CORE-6917 #resolve
- Add annotations for wcslen
- Add kernelspecs.h (complete)
- Add some more entries in driverspecs.h
- Fix a number of annotations
svn path=/trunk/; revision=58204
Add bitmap font for code page 862. Hebrew glyphs are from X11 font
"Schumacher Clean" (see license below), modified by Baruch Rutman
(peterooch*at*gmail*dot*com). CORE-6912.
Copyright 1989 Dale Schumacher, dal@syntel.mn.org
399 Beacon Ave.
St. Paul, MN 55104-3527
Permission to use, copy, modify, and distribute this software and
its documentation for any purpose and without fee is hereby
granted, provided that the above copyright notice appear in all
copies and that both that copyright notice and this permission
notice appear in supporting documentation, and that the name of
Dale Schumacher not be used in advertising or publicity pertaining to
distribution of the software without specific, written prior
permission. Dale Schumacher makes no representations about the
suitability of this software for any purpose. It is provided "as
is" without express or implied warranty.
svn path=/trunk/; revision=58202
- Improve help strings for GCC optimization levels.
- Add new -Ofast and -Og levels.
- Simplify gcc.cmake a bit.
- Use MINGW_TOOLCHAIN_SUFFIX.
svn path=/trunk/; revision=58201
Some keyed event fixes: disable APCs before acquiring the pushlock, close handle on exception, check if a wait was aborted and remove the wait list entry from the list in that case.
svn path=/trunk/; revision=58200
- Add a generic delete function for database objects.
- Do not create Members sub keys for group objects.
- Get rid of the SamDbContainerObject object type.
- Implement SamrDeleteGroup.
svn path=/trunk/; revision=58196
[XDK/DDK] Add missing annotations, move some prototypes to more appropriate headers
[NDK] Fix a number of annotations
[NDK] Use SAL2 dummys from umtypes as well
svn path=/trunk/; revision=58193
- Hide minimized window captions (and enable shell notifications, thus fixing taskbar buttons). This should have been in r57978, but got lost.
- From Andrew Green's GSoC branch. Spotted by Edijus.
CORE-6899 #resolve
svn path=/trunk/; revision=58186
- Italian Translation and fix duplicate MENUITEM in fr-FR.rc, ko-KR, no-NO.rc files. Patch by Ivan Di Francesco.
- Adapt a little bit French translation (by me).
CORE-6908 #resolve #comment Committed in r58185. Thanks !
svn path=/trunk/; revision=58185
[RTL] Use RtlpTimeout for critical sections, which is initialized from the PEB. Implement using the LowMemory keyed event, when allocating the normal event failed.
svn path=/trunk/; revision=58175
- Seperate some usermode only heap functions into their own file and implement dummys in the kernel rtl so that they don't get included in the kernel
- Convert usage of RTL_CRITICAL_SECTION in heappage.c to use of HEAP_LOCK
- Implement A kernel dummy for RtlCallVectoredExceptionHandlers, so we don't put vectored exception handler code into ntoskrnl
- Now we don't have critical section code in the kernel anymore, which wasn't working anyway.
svn path=/trunk/; revision=58173
- SamrAddMemberToGroup: Add the group membership to the user object.
- SamrRemoveMemberFromGroup: Remove the group membership from the user object.
svn path=/trunk/; revision=58161
Add an error icon to the C Runtime termination message box.
Patch by Lee Schröder.
CORE-6886 #resolve #comment Committed in r58157, thanks :)
svn path=/trunk/; revision=58157
Fix the debugging macros introduced in r58132 (for the _FATAL case):
do not use exceptions but instead a breakpoint followed by a process termination procedure (or a bugcheck if we are in kernel mode).
Feel free to improve them !!
[DRIVERS-HAL]
Finish to use the new debugging macros
Part 3/3
svn path=/trunk/; revision=58154
Rewrite DisplayCharacter. The new algorithm is much faster. It uses the VGA mask register to draws up to 8 pixels at once, completely getting rid of the inner loop. Pixel position / mask calculations and the check for opaque vs transparent background are moved out of the loops. The result is a single loop for the y coordinate containing an inlined __outbyte and memory write. A second loop in case of unaligned character writes.
svn path=/trunk/; revision=58148
Apply the following optimizations:
- include ioaccess.h to inline port access, instead of going through hal.
- Make __outpb and __outpw macros rather than stdcall function
- Make SetPixel FORCEINLINE
- Do not switch mode for every pixel we write, instead do it once before doing larger blt operations
- use __movsb instead of manual loop plus READ/WRITE_REGISTER_UCHAR
This noticeably improves performance.
svn path=/trunk/; revision=58147
Add the source login logo .svg file. To make the bitmap, export the .svg file in png, then save the png in Bitmap 24bits.
svn path=/trunk/; revision=58145
Note to people:
~=~=~=~=~=~=~=~
- The msgina (login) logo is made from a .svg file. The copyright line text has font Lucida Sans Unicode (this font allows you to have the right length to the year-dash-separator). DejaVu Sans may also work. I would prefer to use the other logo used as the header for the ReactOS Welcome program (welcome.exe from the CD), but also it is given as a .bmp and I would like to have a .svg or some equivalent format to use it (with little modifications though) as a new login logo (in particular, rescaling it).
- Adjust the boot-logo year, also with the Lucida Sans Unicode font. This was done with Paint from Win2k3 (WARNING!! Not with Windows 7 Paint !!) because it keeps the palette of the opened bitmap, contrary to Windows7 one.
- If somebody has the original boot-logo file (a .svg, ..., something else?), then signal it to me !
svn path=/trunk/; revision=58144
- Use __debugbreak in _assert when user presses ignore in the message box
- Implement __crt_MessageBoxA and use it from _assert and abort
- Rewrite abort and _set_abort_behavior
- Add function headers and set BSD license for my code
- CORE-6594 #resolve
svn path=/trunk/; revision=58143
* For anyone who wants to work only on a certain module (or a set of modules), you can add "project(module)" to the related CMake file. This will generate a solution that contains all the required dependencies needed to compile only that module, without having to open a huge solution with 900+ projects.
* We'll create a wiki page to explain how to use this in detail.
* Dedicated to all VS users. Have fun ;)
svn path=/trunk/; revision=58141
- Fix the debugging macros I've introduced in r58132; in particular do not use while(true); for forbidding the user to continue execution, but instead raise an exception with EXCEPTION_NONCONTINUABLE flag (included when called RtlRaiseStatus).
- Adjust the definition of RtlRaiseStatus (in kernel-mode it is ExRaiseStatus which is used).
svn path=/trunk/; revision=58135
- romanian translation update by Stefan Fulea; CORE-6718
- fix for japaneese cmd translation by Tomoya Kitagawa; CORE-6820
- polish translation update of desk and input;
svn path=/trunk/; revision=58134
Introduce four new debugging macros, ERROR_DBGBREAK, ERROR_FATAL, UNIMPLEMENTED_DBGBREAK, UNIMPLEMENTED_FATAL (and two helpers, __NOTICE and __ERROR_DBGBREAK).
They are designed to display on the debug-log a printf-like user-defined message and to break into the debugger.
- The *_DBGBREAK macros break only into the debugger and allow to continue (they don't hang).
- The *_FATAL macros break into the debugger and then halt the execution.
(Based on an idea of Aleksey Bragin, see the ros-dev mailing list about the while(true) --> assert(false); changes).
[NTOSKRNL]
Use these macros instead of ASSERT(FALSE); introduced in revisions r58110, r58111 and r58112 to replace while(TRUE); halts.
Part 1/3
svn path=/trunk/; revision=58132
- NetUserGetInfo: Add Support for info level 23.
- Convert UserAccountControl flags to user info flags and convert last logon, last logoff and account expires time.
svn path=/trunk/; revision=58130
- Correctly show current time zone name even if no daylight saving settings are set. Patch by Lee Schroeder.
CORE-6803 #resolve #comment Fixed, thanks. Sorry for the long wait.
svn path=/trunk/; revision=58126
- Move "Show window previews (thumbnails)" option to advanced tab, as it doesn't exist in Windows. Patch by Edijs Kolesnikovičs.
CORE-6840 #resolve
svn path=/trunk/; revision=58124
- Plug a memory leak in IDropTargetImpl
- Do not allocate 64x the required space in TaskSwitchWnd_AllocTaskItem
- Properly free and unregister the tray window
- Fix a few more style issues
svn path=/trunk/; revision=58123
Implement SamrQueryInformationUser.UserParametersInformation and set the Parameters attribute when a new user is created.
svn path=/trunk/; revision=58117
In NtGdiStretchDIBitsInternal check early for info dc / mem dc without surface, only allocate memory and copy bits if the caller actually passed bits (they are optional), free the allocation, when an exception happens instead of leaking the memory.
svn path=/trunk/; revision=58116
and in some cases, return an adequate value. In some places however, I add an ASSERT(FALSE); before keeping the while (TRUE); in critical regions.
(Only x86, as well as in r58110). More ASSERTS (in debug mode) or BSODs may appear, instead of having hung threads.
Part 1-bis/2
svn path=/trunk/; revision=58111
Implement DC_hSelectFont, use it from NtGdiSelectFont and GdiFlushUserBatch. Make sure plfnt is != NULL before dereferencing it in "TEXTOBJ_LockText"
svn path=/trunk/; revision=58109
SamrGetAliasMembership:
- Change requrred access right from DOMIN_LOOKUP to DOMAIN_GET_ALIAS_MEMBERSHIP.
- Handle the case properly where a user is not a member of any alias.
- Retrieve an alias members RID correctly.
svn path=/trunk/; revision=58094
Fix untitled Run dialog in some languages. I wonder why it was as it since ages, in some languages but not in anothers.
svn path=/trunk/; revision=58092
- Revert lstrcpynW --> wcsncpy changes from r57471 back to lstrcpynW because this function automatically NULL-terminate strings
(but I wonder if there is a replacement of lstrcpynW which has all its functionalities, in the CRT). This fixes address loading
after editing it in the address bar. Caught by Edijus ;)
- Unicodize a little bit more regedit.
svn path=/trunk/; revision=58090
- In NtGdiGetRandomRgn use the region pointers directly, instead of getting their handles. These regions might not even have a handle.
- Fix last error code
svn path=/trunk/; revision=58089
- Use REG_OPTION_NON_VOLATILE instead of 0 in RegCreateKeyEx.
- Do not rename keys when they have the same name, case-insensitive (as Windows does), fix for r58079 / CORE-6205.
svn path=/trunk/; revision=58087
Exchange fields individually in PDEVOBJ_vSwitchPdev() instead of duplicating a full PDEV structure on the stack. This saves a lot of stack space.
svn path=/trunk/; revision=58081
- Check if the new name equals the old name. In that case don't rename key.
- Fixes regedit accidentaly deleting keys when a key is clicked and the rename field appears
CORE-6205 #resolve #comment finally regedit is usable
svn path=/trunk/; revision=58079
- Add some annotations
- Fix some 64bit issues
- Fix a typo
- reduce ENUM_RECT_LIMIT to 32, which is more than enough and doesn't waste as much stack space
- remove some unused macros
svn path=/trunk/; revision=58076
- Enable the call to NetUserAdd.
This enables us to create new user accounts. A lot of stuff is still missing, so don't be surprised if something unexpected happens. It will be fixed soon.
Happy new Year!
svn path=/trunk/; revision=58072
Pass a PUNICODE_STRING instead of a LPWSTR (and unused BufferSize) to FindMatchingCreateItem(), make use of the Length parameter instead of trusting on NULL termination when searching for path separators.
svn path=/trunk/; revision=58065
Hack the hack a bit more. Should fix annoying warnings about possible uninitialized variables. Also generates smaller code, by getting rid of a stray relative jump instruction.
svn path=/trunk/; revision=58057
Fix value of MmUserProbeAddress, which is MI_USER_PROBE_ADDRESS (0x7FFF0000) and not MI_HIGHEST_USER_ADDRESS (0x7FFEFFFF)!
svn path=/trunk/; revision=58049
In NtAllocateVirtualMemory, when MEM_RESET is passed, silence the DPRINT and also return STATUS_SUCCESS, since MEM_RESET is only an optimization (tells the memory manager that the contents shouldn't be written to the pagefile) and it won't hurt (except the performance) to pretend success. On the other hand paging out is probably never done anyway due to broken Mm.
This fixes spamming the log by Firefox. It does not fix the hangs that firfox suffers from.
svn path=/trunk/; revision=58047
Apply Wine commit "wpp: Fix path separator on Windows platforms." by Hermès Bélusca-Maïto (9dbd1969fca9c31155c6d85ea8129d249467926b)
Indeed, Windows systems use ':' as part of paths and therefore, use another separator character.
This enables compiling ReactOS from source files located in a different disk partition as the generated (output) files.
CORE-6442 #resolve #comment Committed in r58044.
svn path=/trunk/; revision=58044
- Apply Wine commit "widl: Print large enum constants in hex." by Alexandre Julliard (756d33ddae316c71b1241e7c8684c07cd82b9e05)
- Remove some minor differences to Wine
svn path=/trunk/; revision=58039
Store an objects relative id in the database object. This will later be used to protect pre-defined objects from being deleted.
svn path=/trunk/; revision=58038
SamrQueryInformationDomain (DomainGeneralInformation/DomainGeneralInformation2): Count the number of Aliases, Groups and Users.
svn path=/trunk/; revision=58036
- Start implementing a library which is used to parse USB configuration descriptors and construct KSFILTER_DESCRIPTOR structure, which is used with the kernel streaming driver (ks.sys)
- The library will be used in USBAUDIO driver
svn path=/trunk/; revision=58033
* Correct a typo that led to incorrect Capacity value (shown in the disk properties).
* Correct some colors inversion in the Free/Used disk space chart.
* Brought to you by Victor Martinez Calvo.
CORE-6838 #resolve #comment Committed in r58030. Gracias ;)
svn path=/trunk/; revision=58030
- Do not close Notepad when 'Save As' dialog is canceled (behaviour correction, makes it as under Windows). Patch by Peter Hater.
- Temporarily fix saving with invalid file name --> now return an error, don't save the file, reset the internal file name to "" (untitled) and keep Notepad opened.
This is due to the lack of file name validation when calling Save-As file box (should be done by default, but it is not under ROS).
CORE-6836 #resolve #comment Committed in r58024. Thanks :)
svn path=/trunk/; revision=58024
* memcpy should behave exactly like memmove for Windows compatibility.
CORE-6833 #resolve #comment Committed in r58021.
svn path=/trunk/; revision=58021
- Move PcBeep function declaration to a better header.
- Resuscitate OptionMenuCustomBootReactOS() from revision r52491, update it to match recent changes in freeldr as well as making it using new boot method, and reuse ConstructArcPath.
Why I'm doing this ? Because it can be useful to enter personalized boot options by hand at boot time rather than being obliged to edit freeldr.ini. This needs care, not brutal deletion.
svn path=/trunk/; revision=58020
* Fix allocating the required bytes in SetupGetFileCompressionInfoW. Brought to you by Peter Hater (7element {at} mail [dot] bg)
CORE-6815 #resolve #comment Committed in r58018. Cheers ;)
svn path=/trunk/; revision=58018
- Include a netapi32.h in most files and move the other includes into netapi32.h.
- Add functions to open account and builtin domains directly and use these functions.
svn path=/trunk/; revision=58017
Export __crtLCMapStringW and correct __crtLCMapStringA: their prototypes are :
int CDECL __crtLCMapStringW(LCID lcid, DWORD mapflags, const wchar_t *src,
int srclen, wchar_t *dst, int dstlen, unsigned int codepage, int xflag)
and
int CDECL __crtLCMapStringA(LCID lcid, DWORD mapflags, const char* src,
int srclen, char* dst, int dstlen, unsigned int codepage, int xflag)
Needed by SVN.
svn path=/trunk/; revision=58016
- Add Advanced tab in task bar properties to support ReactOS-specific options (currently: show seconds in tray clock). Patch by Edijs Kolesnikovics.
CORE-5410 #comment Patch committed, thanks! Awaiting the updated save-to-registry functionality.
svn path=/trunk/; revision=58011
Improve the way we are doing the timing: When the timeout is > 0, we refresh its display. When it becomes == 0, we exit the waiting-loop and do the right action (start the default OS, ...). However, when we press a key, it is set to -1 by convention, that means, no timeout. The timeout display is then cleared ONLY ONCE since there is no need to refresh it each time we do a waiting-loop-turn when the timeout is == -1.
svn path=/trunk/; revision=58010
- Show the selected boot options as NTLDR does (in light blue under the "For troubleshooting... press F8" message).- The different safe modes are now mutually exclusive, as they must be.
- They can be coupled to the "boot logging", "vga mode" and "debugging" modes, enlarging possibilities (contrary to NTLDR. However you'll remark that these combinations are allowed when set with msconfig...).
- Since the safe modes come by default with boot logging, don't show "Boot Logging" when one of the safe modes is selected, but instead show the corresponding safe mode name.
- Modify the needed functions.
svn path=/trunk/; revision=58008
- Lee Schroeder: Replaces the current fontview icon with the icon found in the tango icon set (apps/preferences-desktop-font.png). This creates better consistency since that is where the majority of the icons come from.
CORE-6827 #resolve
svn path=/trunk/; revision=58007
Allow VideoMode to be set via decimal, hexadecimal, ... number representation in bootcd/livecd/freeldr.ini (see the Brian Palmer's FREELDR.INI example).
svn path=/trunk/; revision=58005
- Menu titles --> Menu headers (see r57842).
- In the same way we introduced menu headers, introduce menu footers, therefore we add the possibility, in mini-tui,
of showing or hiding the "For troubleshooting and advanced startup options for ReactOS, press F8." message, only
useful for the main menu (the one which shows the list of available OSes installed on the system).
svn path=/trunk/; revision=58001
Add some spacing (not too much) between the screen borders and the "Safe Mode" and the ReactOS version string labels, in Safe Mode.
svn path=/trunk/; revision=58000
Add a suitable 'Fixedsys' TTF Unicode font, called 'Fixedsys Excelsior v3.01'.
Released into the Public Domain.
By Darien Gavin Valentine,
http://www.fixedsysexcelsior.com/
svn path=/trunk/; revision=57998
- Implement NetUserGetInfo.
- Implement level 10 in BuildUserInfoBuffer.
- Zero-initialize the user info buffer after allocation.
svn path=/trunk/; revision=57997
- Remove "Properties" from the task bar property sheet title, as it is automatically appended. Patch by Edijs Kolesnikovics
CORE-6826 #resolve
svn path=/trunk/; revision=57993
* Add basic support for the new MSVC build handling that comes with the upcoming RosBE 2.1.
* This needs to be inspected further to make sure it matches our current setup.
svn path=/trunk/; revision=57990
1) Use of CheckDlgButton() and IsDlgButtonChecked() rather than SendMessage, it will improve code readability.
2)Removes an useless initialization into perfpage.c
3) GetWindowLong/SetWindowLong were replaced with GetWindowLongPtr/SetWindowLongPtr, but GWL_WNDPROC and GWL_ID were not with GWLP_WNDPROC and GWLP_ID.
4) Corrected some comment texts.
Patch by Carlo Bramini
CORE-6367 #resolve
svn path=/trunk/; revision=57989
- Use in some places ZeroMemory instead of memset(..., 0, ...).
- Fix way of initializing the STARTUPINFO structure before call to CreateProcess: in particular, when using the wShowWindow field, also use the STARTF_USESHOWWINDOW flag.
[USERINIT]
- Fix way of initializing the STARTUPINFO structure before call to CreateProcess: in particular, when using the wShowWindow field, also use the STARTF_USESHOWWINDOW flag.
- This fixes CORE-5735.
CORE-5735 #resolve #comment Command-line launching fixed in revision r57986.
svn path=/trunk/; revision=57986
- Add support for task bar theming
- Add notification area support
- Add support for window flashing
- Hide minimized windows
- Based on Andrew Green's GSoC branch (3/3)
svn path=/trunk/; revision=57978
- Run file dialog in its own thread
- Correctly check MapWindowPoints return value
- Fix a warning
- Based on Andrew Green's GSoC branch
svn path=/trunk/; revision=57977
- Fix a bug during socket closure that could cause buffered data to be lost
[LWIP]
- Only signal errors when the packet queue is empty to avoid dropping received data
- Use pbuf_copy_partial to copy pbuf chains instead of doing it manually which had a bug
CORE-6655 #resolve CORE-6741 #resolve
svn path=/trunk/; revision=57975
Fix a bug in the GDI pool code that could lead to list corruption and a failed ASSERT, when an app allocated a large number of DCs or brushes, like AbiWord with more than 1 document open. Also add a few more ASSERTs.
svn path=/trunk/; revision=57973
- Fork netapi32.dll because most of its exported functions are interfaces to services that wine will not implement.
- Move user and group functions to separate files and add some stubs.
- Update README.WINE.
svn path=/trunk/; revision=57972
- Rewrite NtUserGetAtomName, make it return the length in WCHARs instead of bytes, as it is supposed to be. Fix a buffer overrun, when the caller passes a too large UNICODE_STRING buffer. Probe the output buffers before accessing them. Makes sure the target string is always NULL terminated, even if the buffer is too small for the whole name.
- Remove NULL termination code from IntGetAtomName, since that is already done by RtlQueryAtomInAtomTable
svn path=/trunk/; revision=57970
- Upgrade the files architecture to the classic one : lang-directory for language resource files, rsrc.rc and <app_name>.rc for main resource files.
- Convert remaining language resource files to UTF-8 without BOM.
- Add version information to the program - Patch by Lee Schroeder.
CORE-6821 #resolve #comment Committed in r57963. Thanks :)
svn path=/trunk/; revision=57963
- Update RedHat Liberation Fonts to 2.00.1
- Update and add some DejaVu Fonts to 2.33
- Update GNU FreeFonts to revision 20120503
- Update Tahoma and Symbol Fonts to Wine HEAD
- Sort the fonts documentation files to dedicated directories
svn path=/trunk/; revision=57959
- Use 2 seperate SEH blocks in PsSuspendThread(). This way we don't leak the rundown protection, and don't keep a terminating thread suspended, when KeSuspendThread() throws an exception. Also only catch STATUS_SUSPEND_COUNT_EXCEEDED exceptions in the SEH filter.
svn path=/trunk/; revision=57958
- Add some missing return value checks
- Add some annotations
- Fix format specifiers
- Fix variable scope conflicts
- Fix possible closing of a NULL handle
- Use ObCloseHandle instead of ZwClose
svn path=/trunk/; revision=57954
- Add/improve some annotations
- Improve code in EngpGetPDEV and NtGdiGetRegionData
- Simplify SEH exception handling
- Fix possible NULL pointer dereferences
- Fix wrong variable usage in UnpackParam
- Fix a possible bufer overrun
- Check return value of ExAllocatePool and RtlCreateUnicodeString
- Fix & vs && typo
All detected with VS11 analyse
svn path=/trunk/; revision=57948
- services.exe is indeed a GUI program, so build it as such.
- Check the return value of RegisterServicesProcess and thus allow only one instance of services.exe to be launched.
CORE-6123 #resolve #comment Fixed by r57946.
svn path=/trunk/; revision=57946
- Fix a calculation bug in the heap code
- Fix pointer arithmetics
- Fix artihetic overflows
- rename some global items to not conflict with local names
- Silence some more warnings from VS analyze
svn path=/trunk/; revision=57936
- Effectively reboot instead of exiting the bootloader and hanging.
- Do a cold reboot instead of a (soft) warm reboot, to let the user be able to choose another boot support or do anything else, instead of directly rebooting to the bootloader. Rename the concerned function accordingly.
svn path=/trunk/; revision=57935
Completely remove the special blue "setup-ui" appearing when starting the bootcd, in this way revert a part of r57842 and r57843.
Requested by some people. Now we keep the NTLDR-like boot ui.
svn path=/trunk/; revision=57933
Start rewrite sal.h
This will drop support for WDK's prefast, since it's impossible to maintain both sal version 1 and 2 and the old stuff is quite unreliable anyway. Only MSVC 11+ analyze with attributes for SAL will work.
abragin: "\o/ for throwing away some legacy" ;-)
svn path=/trunk/; revision=57932
Fix BSOD when starting ReactOS in Safe Mode :
NEVER dereference a NULL pointer in a DPRINT or you are dead !! (here, the possible NULL pointer is GetW32ProcessInfo()->prpwinsta).
More precisely, you obtain:
<NTOSKRNL.EXE:112296 (ReactOS/ntoskrnl/ke/i386/traphdlr.c:1246 (@KiTrap0EHandler@4))>
<NTOSKRNL.EXE:110154 (ReactOS/ntoskrnl/ke/i386/trap.s:0 (_KiTrap0E))>
<NTOSKRNL.EXE:142926 (ReactOS/lib/sdk/crt/printf/streamout.c:644 (streamout))>
<NTOSKRNL.EXE:140631 (ReactOS/lib/sdk/crt/printf/_sxprintf.c:61 (_vsnprintf))>
<NTOSKRNL.EXE:12736a (ReactOS/lib/rtl/debug.c:85 (vDbgPrintExWithPrefixInternal@24))>
<NTOSKRNL.EXE:12751c (ReactOS/lib/rtl/debug.c:209 (DbgPrint))>
<win32k.sys:55034 (ReactOS/win32ss/user/ntuser/sysparams.c:451 (SpiGet))>
<win32k.sys:56bc7 (ReactOS/win32ss/user/ntuser/sysparams.c:963 (@UserSystemParametersInfo@16)
svn path=/trunk/; revision=57923
Improve functionnality of the shutdown utility :
* Added extra arguments based upon Windows 7.
* Added options to CommandLineOptions function.
* Simplify the PrintResourceString() function.
* Removed PrintUsage() since PrintResourceString() does a more flexible job at printing strings out.
* Added clear error messages when there is a problem. This makes it a bit easier to determine if there is a problem.
* Convert all the C++ comments to C.
* Since each version of Windows has a different upper limit for the length of the comment, a work around solution is used that first checks the length of the comment based upon the version of Windows that the user is currently using.
* Added support for hibernation mode, but the function SetSuspendState() is only a stub right now.
* In cases where the shutdown/restart operation is done remotely, now the proper permission is granted if the user has the proper permission.
* Added the REASON structure, which contains all the reason code combinations required for giving the proper reason code to the computer.
* Added bare bones basic support for the GUI option (/i). However, until the required features are added to ReactOS, the GUI won't do anything.
* Convert the program to full UNICODE support.
CORE-6613 #resolve #comment Committed in r57916. Thanks :)
svn path=/trunk/; revision=57916
- Use a list of possible operating system method loading to load ReactOS / ROS-Setup / Windows / Linux or booting from a disk / partition ..., instead of using a (huge) if () else if () ... instruction.
- Modify the corresponding functions to take into account for this new functionality.
- Allow having a live + boot cd if needed, in a elegant way.
svn path=/trunk/; revision=57899
SetupDiLoadClassIcon: Return the handle to or index of the default icon (Other devices icon) if no valid ClassGuid was provided by the caller.
With this patch, the class icon of 'Other devices' appears in the device property sheets in the device manager.
svn path=/trunk/; revision=57884
* Reduce palette window size.
* Fix some Japanese text.
* Brought to you by Hirofumi Katayama.
CORE-6802 #resolve #comment Committed in r57870. Thanks ;)
svn path=/trunk/; revision=57870
* Update some shell related headers.
[UUID]
* Add some GUIDs to shguids.c.
[MAIN.CPL]
* Remove initguid.h inclusion and link to uuid.
svn path=/trunk/; revision=57857
Reintroduce the long-awaited boot-logo fading-in, removed since ReactOS 0.3.3 due to bootvid / inbv refactoring. Code taken from the old (ie. pre-ReactOS 0.3.3) bootvid.
svn path=/trunk/; revision=57855
Improve the overall look:
- Augment the timeout of the livecd from 3 seconds to 5 seconds.
- In "Mini"-UI (i.e. NTLDR-like UI), add a newline between "Press ENTER" and the time counting.
- Have a real "Setup"-UI (i.e. make it really work).
- Add the possibility to change the title of menus.
svn path=/trunk/; revision=57842
- make some functions locale aware
- add implementation of ___mb_cur_max_func and _isctype_l
[MSVCRT20]
- keep up with msvcrt
[MSVCRT40]
- ditto
svn path=/trunk/; revision=57838
- LsarQueryAuditEvents: Do not fail if the PolAdtEv attribute does not exist and return proper inormation to the caller.
- Implement LsarSetAuditEvents.
svn path=/trunk/; revision=57837
- Add a trusted flag to the lsa object type.
- Inherit the trusted flag from the policy object when an account or secret object is created or opened.
- Set the trusted flag for a policy object in LsaIOpenPolicyTrusted.
svn path=/trunk/; revision=57821
- Only dissolve a stack of cards if they're of the same color
- Allow easier recognition of suit in concealed cards
- Patch by Marcel Leyendeckers, m dot leyendeckers at gmx dot de
CORE-6808 #resolve #comment Committed. Thanks!
svn path=/trunk/; revision=57815
- Set USBD status when a device is successfully configured and transferred length for vendor class requests
- Add error msg when copying of interface descriptor fails
svn path=/trunk/; revision=57813
- Add check if interface descriptor was parsed
- Correct error message
- Set device idle after a configuration has been selected
- CORE-6574
svn path=/trunk/; revision=57805
- Set ParameterValue when NdisReadConfiguration is successful
- Fixes crash when Belkin F5D8053 N Wireless USB Adapter is initializing
- Introduced in rev 5338 (31/07/2003)
svn path=/trunk/; revision=57803
- Handle devices which provide indicate a serial number but not provide one
- Fixes device installation initialization for F5D8053 N Wireless USB Adapter
svn path=/trunk/; revision=57799
- Remove the right element from the list when uninitializing a file lock
- Do not modify the content of the table range while looping on its elements
- Fix rebuilding the table when unlocking a shared range
Thanks go to Pierre for the hint for where to look at
CORE-6615 #comment fixed by r57798 #resolve
CIRE-6535 #comment fixed by r55798 #resolve
svn path=/trunk/; revision=57798
- Skip other spaces preceding a command option instead of increasing the reading pointer of a constant (should fix KDSERIAL).
- Use a macro for representing length in chars of constant strings instead of putting magic values.
svn path=/trunk/; revision=57787
- Fix object deletion.
- Add a registry wrapper function that enables us to delete registry keys which have already been opened.
- Make LsarDelete call LsarDeleteObject because otherwise object deletion would not be possible.
svn path=/trunk/; revision=57784
Fixes COM port debugging output with Virtual PC 2007. See r57781 for more details.
CORE-4247 #comment Fixed also in r57782.
svn path=/trunk/; revision=57782
- Use cportlib for COM port facilities. Fixes COM port debugging output with Virtual PC 2007.
- Remove now unneeded header file.
CORE-4247 #comment Fixed in r57776, r57777, r57780 and r57781. #resolve
svn path=/trunk/; revision=57781
GetByte --> Wait for data (with timeout) and get it if available.
PollByte --> Check for data, get it if available, and return immediately.
svn path=/trunk/; revision=57780
- Make use of the previously committed cportlib code. This works great on VPC 2007 :)
- Comment the unused Rs232PortPollByte function.
- In hardware.c, add useful comments and two macros holding the maximum number of COM and LPT ports.
svn path=/trunk/; revision=57777
- Do a synthesis of all the code I've found concerning initializing COM ports (in kdcom / kddll / kdbg in the kernel and in freeldr). The "// Windows" comments concern the original code for testing the existence of COM ports, inside cportlib, whereas "// ReactOS" comments concern the equivalent existing in our kdcom/kddll/kdbg etc...
- CORRECT THE VIRTUAL PC COM PORT INITIALIZATION !! The problem was, that the current COM port existence test fails on VPC whereas it works well in all other virtual machines. Therefore :
* I keep the existing testing code in a function called ComPortTest1, this works everywhere but on VPC ;
* I add a very basic COM port existence test function called ComPortTest2, which basically looks for the scratch register, this works everywhere including on VPC.
* The COM port existence test consists in the two tests above.
svn path=/trunk/; revision=57776
- Implement LsarDeleteObject.
- Store the object name in the database object in order to be able to delete the objects key later.
svn path=/trunk/; revision=57769
- Add enumeration of user rights to LsarEnumerateAccountRights.
- Use RPC_UNICODE_STRING instead of UNICODE_STRING in the privilege lookup code.
svn path=/trunk/; revision=57767
- (Re-)implement acquiring/releasing driver-loading and unloading privilege (Note: the two privileges have the same name SE_LOAD_DRIVER_NAME).
- Rework the SCM initialization code. In particular, acquiring the starting lock MUST BE done BEFORE starting the RPC server, otherwise, a program can try to start a service (and during this operation, the global start lock is acquired by the SCM) DURING the initialization of SCM, and in particular between the call to ScmStartRpcServer and ScmAcquireServiceStartLock. The SCM would then see an already-acquired start lock at this point and would fail to initialize. ==> This commit corrects this behaviour.
svn path=/trunk/; revision=57754
When filling LoadOptions, strip the '/' commutator symbol in front of each option. Now we fully comply with Windows when holding LoaderBlock->LoadOptions, and the content of the registry value SystemStartOptions under HKLM\CurrentControlSet\Control display the options without their / symbol.
[WIN32K]
Now, the SystemStartOptions value doesn't contain / anymore.
svn path=/trunk/; revision=57752
I think this history deserves champagne and forceful warning comments, which have been added.
Simple solution: Don't blindly sync Wine code!
#resolve #CORE-6495
svn path=/trunk/; revision=57747
NetLocalGroupEnum: Add code to return information about a single alias to the caller.
Now, the user account control panel enumerates the built-in groups. ;-)
svn path=/trunk/; revision=57700
Resize the dialogs to fit the ReactOS image, and add the logo to the "Please wait..." dialog
(which is displayed when a user is logging-on or quitting his session and the "Save your parameters..."
message is displayed). Now ReactOS looks perfect.
svn path=/trunk/; revision=57697
- Implement usage of alpha cursors (does not work yet due to a bug in windres)
- Directly use bitmaps passed to NtUserSetCursorIconData instead of copying them
svn path=/trunk/; revision=57695
- Add test showing that a bitmap created using CreateDIBItmap can be selected into a DC despite a different bits depth
svn path=/trunk/; revision=57693
- Start using CURSORDATA structure for passing data from user to kernel mode.
- Start using the proper CURSOR structure for win32k internal represesentation of cursor and icons
- Create the alpha bitmap in user mode instead of kernel mode.
svn path=/trunk/; revision=57692
- Work only with BITMAPINFO in GreGetDIBItsInternal and convert the structure back to BITMAPCOREINFO if needed.
- Set the rgbReserved member to 0 when getting palette of a DIB section
svn path=/trunk/; revision=57691
NetLocalGroupEnum: Make sure that built-in aliases are retrieved once and only one alias is enumerated upon each consecutive call. Retrieve name and comment for each alias.
svn path=/trunk/; revision=57684
SamrQueryInformationAlias: When retrieving general information, return a member count of zero if the members sub key does not exist.
svn path=/trunk/; revision=57682
- Implement LookupIconIdFromDirectoryEx.
- Implement CopyImage
- Finish Implementing LoadImage
- Implement various functions.
[WIN32SS]
- Add a first working implementation for LR_SHARED cursors and icons.User32 support is there, but more work is needed in win32k.
svn path=/trunk/; revision=57672
- Move the ReactOS-specific workaround described in http://jira.reactos.org/browse/CORE-6611 and http://jira.reactos.org/browse/CORE-4620
from kernel32 to ntdll for using it by RtlGetVersion, in order to :
* still having it available in user-mode only,
* and having VerifyVersionInfo (and its Rtl counter-part) working properly.
[RTL/NTOSKRNL]
- Code formatting.
- Use a defined-constant instead of a hard-coded value.
svn path=/trunk/; revision=57662
Fix report of OS version: we report us as Windows Server 2003 SP1, in a consistent way, i.e. our product type is ServerNT and not WinNT (workstation).
However, for applications which would need to see us as a workstation OS, a workaround exists and can be activated manually by the user (by default it is not).
See http://jira.reactos.org/browse/CORE-4620 for a description of the workaround.
CORE-6611 #comment In r57661, our version is reported consistently.
svn path=/trunk/; revision=57661
- Remove InputDesktopHandle variable because it is a handle and that means that it can only be used in the context of winlogon
- Fixes NtUserOpenInputDesktop which never really worked
svn path=/trunk/; revision=57659
- Use ONEPARAM_ROUTINE_GETCURSORPOS instead of ONEPARAM_ROUTINE_GETCURSORPOSITION in GetCursorPos
- Correctly check if the current thread belongs to the active desktop
svn path=/trunk/; revision=57657
- Properly handle deletion of cursor/Icon objects
- Gracefully change object owner when process closes
- Remove useless check in a macro
svn path=/trunk/; revision=57641
- Double the size of the user handle table in order to allow the test suite to complete despite the horrible object leaks
svn path=/trunk/; revision=57624
- Fix WARN macros
- Keep a list of the PROCESSINFO of all the running processes
- Count how many handles a process owns per type in PROCESSINFO
- Improve the debug output when we run out of user handles to show all handle counts per process
svn path=/trunk/; revision=57623
- From version 9.0, mesa doesn't ship the glu library anymore. It is maintained in another tree.
- Introduce glu 9.0 to ReactOS code base.
svn path=/trunk/; revision=57621
- Use LookupAccountNameW to retrieve the account SID when a user tries to log-on to a computer.
- Little clean-up of LogonUserW.
This is the first step to get rid of hard-coded logon stuff.
svn path=/trunk/; revision=57601
- Implement HidP_GetUsageValue
- Add hack to HidParser_GetScaledUsageValueWithReport
[MOUHID]
- Add partial support for absolute pointing devices
- Tested in VBOX 4.1.22 with absolute pointing devices
- Needs more work in win32k
- Core 6553
svn path=/trunk/; revision=57595
- Change the account and domain names in the WELL_KNOWN_SID type to AccountName and DomainName.
- Implement a function to look-up fully qualified well-known names (for example "NT AUTHORITY\LOCAL SERVICE"). This kind of look-up was still missing from LsapLookupNames.
- Ensure that all well-known SIDs returned by look-up functions are copied from the list of well-known SIDs, because RPCRT4.DLL frees the SID buffers after copying the SIDs to the caller. This will corrupt the well-known SID list if pointers to the original SIDs are passed.
- Check all memory allocations.
svn path=/trunk/; revision=57581
Add setupapi.dll to the KnownDlls list.
CORE-6732 #comment Quake II setup fixed in rev 57576. Patch by Lee Schroeder. #resolve
svn path=/trunk/; revision=57576
- Add well-known account names "LOCALSERVICE" and "NETWORKSERVICE" as aliases of "LOCAL SERVICE" and "NETWORKSERVICE". They can be found on a german Windows XP.
- Remove a part LsaLookupSids that returned "UNKNOWN\Test" for unknown SIDs. Was used for tests only.
svn path=/trunk/; revision=57568
LsapLookupIsolatedNames: Add a domain entry to the referenced domain list for an unknown/unidentified domain. Its SID is the SID of the account minus the last sub-authority and its name is an empty string.
svn path=/trunk/; revision=57557
- Use COM for calling SHBrowseForFolder.
- Allow choosing a directory for downloading.
- Correct a typo in the name of a function.
svn path=/trunk/; revision=57553
- Pre-implement InitiateSystemShutdownExW.
- Cosmetic formatting code (by me).
Patch by Lee Schroeder. See CORE-6613 for more details.
CORE-6613 #comment Committed shutdown.c.
svn path=/trunk/; revision=57546
Correct the SP number (==> SP1) we are reporting (currently SP1 of Windows Server 2003). The SP6 thingie was a remainder from our previous target, Windows NT 4 SP6.
svn path=/trunk/; revision=57543
Use a temporary buffer to parse the program's arguments, instead of using (and thus overwriting) the _a/wcmdln variable.
Fix msvcrt_apitest:CommandLine test.
CORE-6717 #comment Fixed in revision r57536. #resolve
svn path=/trunk/; revision=57536
- Assume that the provided buffer size is maximal in GetDIBits
[WIN32K]
- Probe buffer before writing to it
- Fail DIB Section creation if palette creation failed
svn path=/trunk/; revision=57535
- Fix a bug in RtlAreAnyAccessesGranted
- Fix an MSVC warning in RtlEqualPrefixSid
[CRT]
- Fix an MSVC warning in _ui64tow_s
[NTOSKRNL]
- DPRINT the filename when MmLoadSystemImage fails
- Fix a few MSVC warnings
svn path=/trunk/; revision=57518
Implement lookup of accounts in the account domain of the SAM database. The advapi32 security winetest shows proper domain and account names for the administrator and guest SIDs.
svn path=/trunk/; revision=57517
Implement SamIFree_SAMPR_RETURNED_USTRING_ARRAY and SamIFree_SAMPR_ULONG_ARRAY.
These functions are used to free structures that were allocated by calls to Samr-functions.
svn path=/trunk/; revision=57516
In NtQuerySystemEnvironmentValue :
- Use a fixed-length buffer size for holding a temporary variable.
- Correct the pool TAG.
svn path=/trunk/; revision=57514
- Store the names and SIDs of the built-in and account domains in global vaiables.
- Move the lookup code for isolated names to a separate function.
svn path=/trunk/; revision=57511
Use a struct do define the combined DIBSECTION/BITMAPINFO buffer in CreateCompatibleBitmap, to make it more clear how it works. See CID 716216
svn path=/trunk/; revision=57509
Chnge return type of ClientLoadLibrary to BOOL. Previously it returned HMODULE in both the load and unload case, being a meaningless value in the latter case. All users of this function were using it as a boolean parameter only.
svn path=/trunk/; revision=57507
- Make LsarLookupNames, LsarLookupNames2 and LsarLookupNames4 call LsapLookupNames and convert the results where this is needed.
- Split the lookup code in LsapLookupSids into separate functions.
svn path=/trunk/; revision=57503
- Make LFONTs (for some incomprehensible reason called TEXTOBJ in ros) shared objects, as they are supposed to be. Since that conflicts with our totally broken font code that happily messes with the LFONT's contents, an additional push lock is aquired to get exclusive ownership.
svn path=/trunk/; revision=57494
- Tested wine win test_scrollwindow, test_scrollvalidate, test_scrolldc, and test_scroll plus ReactOS API tests. Pass all but three region tests in test_scrollvalidate.
- Dedicated to Víctor Martínez Calvo.
svn path=/trunk/; revision=57491
- Change LsapLookupSids to match LsarLookupSids2 and LsarLookupSids3.
- Implement LsarLookupSids2 and LsarLookupSids3.
- Update the call to LsapLookupSids in LsarLookupSids and convert the translated names list after calling LsapLookupSids.
svn path=/trunk/; revision=57489
- Move most of the code from LsarLookupSids into LsapLookupSids.
- Use the code from LsapLookupNames to handle the referenced domains list in LsapLookupSids.
svn path=/trunk/; revision=57488
- Fix what might finally be the last missing "this->" instances left over from the C++ conversion (used a regex this time)
svn path=/trunk/; revision=57485
- Convert the application's code to UNICODE
- But the regproc.c's code is untouched (apart from the pseudo-error messages issued with fprintf)
svn path=/trunk/; revision=57471
Fixes several bugs in the current implementation of LsapSplitNames and LsapLookupNames. This fixes the crashing lsa advapi32 wintest. LsapLookupNames is still WIP.
svn path=/trunk/; revision=57464
- WCHAR ==> TCHAR in (before a REALLY conversion of regedit into UNICODE)
- Correct some mistakes in displayed strings
- Improve informative / warning / error messages when importing registry files
- Now, try to import each registry file specified in the command line and don't fail miserably when one of them is missing, but continue with the next one (as Windows' regedit does).
Patch by Edijs Kolesnikovičs (2/2) adapted by me.
CORE-6239 #comment Fixed in r57462 with some adaptations. Thanks ;) #resolve
svn path=/trunk/; revision=57462
Move gdi kdbg extension into a separate file, prefix names with ! to match WinDbg extensions more closely (e.g. "!gdi.help"), imlement help, dumpht, handle, entry and eventlist commands.
svn path=/trunk/; revision=57459
- Fix comment headers.
- Improve error-displaying function and add another one (to be used later on).
- Clear a part of code.
svn path=/trunk/; revision=57456
Implement Kdbg CLI callbacks. These callbacks can be registered from external modules and allow to process kdbg commands.
[WIN32K]
Implement kdbg callbacks for gdi: "gdi!dumpht <type>" dumps the handle table, "gdi!handle <x>" dumps information about a handle.
CORE-6650 #resolve
svn path=/trunk/; revision=57454
Get rid of the ugly container object code. Let LsapCreateDbObject and LsapOpenDbObject handle container objects instead.
svn path=/trunk/; revision=57449
- Validate Account SIDs in LsarCreateAccount and LsarOpenAccount.
- LsarOpenAccount: Do not check the granted access of the policy handle.
- LsarOpenAccount: Return the proper status code.
svn path=/trunk/; revision=57448
- Correct command processing following the first command in interactive mode.
- Correct invocation of command-specific help functions.
svn path=/trunk/; revision=57447
- Make srb volatile, since it's assigned inside SEH and referenced in finally
[NTOSKRNL]
- FsRtlTeardownPerStreamContexts: make IsMutexLocked volatile (SEH)
- IoCreateFile: Make SystemEaBuffer volatile (SEH), save status and information in the caller's IoStatusBlock, cleanup and fail when IoCheckEaBufferValidity failed with PreviousMode == KernelMode, too.
- NtLockFile: Move ExAllocatePoolWithTag out of the SEH block. ExAllocatePoolWithTag does not raise an exception by default (unlike ExAllocatePoolWithQuotaTag). Get rid of this SEH block completely and check the return value instead.
- NtQueryDirectoryFile: make AuxBuffer volatile (SEH), again move ExAllocatePoolWithTag out of the SEH block and check return value instead.
IopCaptureUnicodeString: Make Name volatile (SEH)
svn path=/trunk/; revision=57437
Add a "comment" that will notify whoever wants to compile this poc, to move the IoCompleteRequest out of the SEH block.
svn path=/trunk/; revision=57435
- Allocate a capture buffer outside of SEH in NtAddAtom and NtRaiseHardError
- Call ExpGetCurrentUserUILanguage outside of SEH in NtQueryDefaultUILanguage and return the actual result instead of returning PsInstallUILanguageId in case of success.
- Don't allocate a unicode string inside SEH in NtQuerySystemEnvironmentValue, instead use RtlAnsiStringToUnicodeString directly on the callers buffer.
CORE-6624
svn path=/trunk/; revision=57429
- Reset read and write events directly before waiting. Fixes an issue where a read could return 0 bytes if the previous read had read all available data without waiting (thus leaving the event set).
svn path=/trunk/; revision=57424
- Do not access ThreadContext after freeing it in NpfsWaiterThread
- Correctly update Vcb->EmptyWaiterCount when terminating in NpfsWaiterThread
- Fix more MSVC warnings, lots of DPRINTs
svn path=/trunk/; revision=57421
- Fix return status if no listening server found in NpfsCreate
- Do not use obsolete function MmGetSystemAddressForMdl
- Fix a few MSVC/GCC 4.7 warnings
- Correctly handle device/root FCB allocation failure
- Fix type of NPFS_CCB::RefCount
svn path=/trunk/; revision=57419
- Revert rev.57413. Indeed, when one boots ReactOS without debugging (selecting the first entry in the boot menu), the COM1 symbolic link (corresponding to \Device\Serial0) is created automatically by the serial driver. However, when one boots in debugging mode (selecting the second entry -- ReactOS (Debug) --), then for some reason the COM1 link isn't created (albeit \Device\Serial0 is). Therefore the real problem is elsewhere. Seems that it has something to share with http://jira.reactos.org/browse/CORE-4263
svn path=/trunk/; revision=57414
- HACK: Add artificially a symbolic link to the first COM port, because it happens for
whatever reason that the serial.sys driver doesn't create it when it creates the Serial0
device. When this functionality is repaired, the HACK shall disappear.
svn path=/trunk/; revision=57413
- Revert rev.57400 only for this file since CID 701342 was a false-positive from Coverity (because Coverity doesn't know what the aim of the POOL_RAISE_IF_ALLOCATION_FAILURE is).
svn path=/trunk/; revision=57411
- Implement LsaRemovePrivilegesFromAccount.
- Add stubs for LsaGetRemoteUserName, LsaLookupPrivilegeDisplayName, LsaOpenPolicySce and LsaSetInformationTrustedDomain.
- Fix some issues in the spec file.
svn path=/trunk/; revision=57410
- Reworking the Find* APIs in kernel32. Fixes some kernel32_winetest:file failures.
- The FindExInfoBasic information level for FindFirstFileEx, is also implemented.
- Update ndk and psdk headers regarding to this work.
CORE-6623 #comment Committed in rev.57407. #resolve
svn path=/trunk/; revision=57407
- Implement the PolicyDnsDomainInformation class of LsarQueryInformationPolicy and enable the PolicyModificationInformation class.
- Add required attributes to the initialization code.
svn path=/trunk/; revision=57397
- Implement most missing information classes of LsarQueryInformationPolicy.
- Add initialization code for new attributes of the policy object.
- Implement LsarQueryInformationPolicy2 and LsarSetInformationPolicy2.
svn path=/trunk/; revision=57391
- Remove duplicated definitions from ntddk.h
- Fix annotations in ntifs.h, wdm.h, ntgdi.h, winddi.h
[NDK]
- Declare most of the heap functions only for user mode
- Remove duplicate RtlGetNtGlobalFlags()
svn path=/trunk/; revision=57389
- Set the default quota limits when the LSA database is created.
- Implement the PolicyDefaultQuotaInformation class of LsarQueryInformationPolicy.
svn path=/trunk/; revision=57383
- Convert the application to Unicode.
- Implement "net start" and "net stop" commands that use the service manager APIs.
svn path=/trunk/; revision=57375
Unmap the desktop heap in IntFreeDesktopHeap. Fixes leaking session space memory, when creating and deleting desktops.
I would revert the "hackfix" from r57348, but for me changing MiPteToAddress to MiAddressToPte looks more like a legit bug fix, rather than "increasing system view space to 36MB instead of 20MB". But I'm probably missing something. In that case, Alex, please revert your hack, it shouldn't be needed anymore.
CORE-6688 #resolve
svn path=/trunk/; revision=57367
- Fix overlapped I/O error handling for pipe transport
- ROS now works with Windows npfs.sys again (CORE-2198)
- Wine has this completely reworked, so this ros-diff should finally disappear with the next sync
svn path=/trunk/; revision=57365
- Separate condition calculation in RtlVerifyVersionInfo into an extra function. Patch by Hermès Bélusca (2/X)
- Don't make the Hungarian more ridiculous than it needs to be
CORE-6611
svn path=/trunk/; revision=57355
- Skip all qmgr tests because they cause timeouts on testbot. This should hopefully be fixed soon and they can be re-enabled.
CORE-6536
svn path=/trunk/; revision=57353
- hungarian calc translation by Nagy Tibor(xnagytibor at gmail dot com); CORE-6511 #resolve #comment Commited. Thanks for your work
- dutch rapps translation update by stijnsanders (stijn at yoy dot be); CORE-6540 #resolve #comment Commited. Thanks for your work
svn path=/trunk/; revision=57350
- Correctly show the amount of available memory in the serial debug log. Patch by Hermès Bélusca (2/2)
CORE-6635 #comment Committed. Thanks. #resolve
svn path=/trunk/; revision=57333
make sure at least one rect is allocated in REGION_PtsToRegion. Fixes failed assertion when starting pacman4000 (the game itself doesn't work though)
CORE-5431 #resolve
svn path=/trunk/; revision=57329
This patch fixes various things, from Coverity code defects to conversion
warnings :
- CID 715948 (logically dead code @ services/rpcserver.c)
- try to fix CID 716332/3 (resource leaks) by rewriting the ScmReadString
function (@ services/config.c)
- zero out the freshly allocated memory (@ services)
- try to fix CID 716126/7/8 (untrusted value as argument @
advapi32/services/sctrl.c)
Fix also some "size_t to DWORD" warnings on x64 build (@
advapi32/services/scm.c).
Patch by Hermes BELUSCA - MAITO.
Fixes CORE-6606.
svn path=/trunk/; revision=57328
Allow passing NULL to EngFreeMem, like on Windows.
Fixes mode switching with cirrus driver on qemu.
CORE-6335 #resolve
svn path=/trunk/; revision=57325
Fix display of formerly trucated info text in German language.
Patch by Robert Naumann (robert dot naumann at ovgu dot de)
CORE-6218 #resolve
svn path=/trunk/; revision=57323
Rewrite FAT bootsector. It's written to eventually replace FAT12, FAT16 and FAT32 bootsectors using the same code. Currently only used for FAT16. The complete (!) code to load freeldr has been put into a single 512 byte sector. CHS support is not implemented atm, but there's plenty space left (41 bytes). Fixes boot on harddisks smaller than 513 MB.
CORE-6610 #resolve
svn path=/trunk/; revision=57319
* Don't add the whole absolute path to GCC for preprocessing when creating RC dependency files in Ninja builds.. that complicates the already complicated quoting needs here.
svn path=/trunk/; revision=57302
- Make a copy of the data type information instead of writing to the (read-only) .rsrc section. Already applied upstream.
CORE-6645 #comment Should be fixed. Can you retest with HEAD please?
svn path=/trunk/; revision=57298
Implement RtlpSafeCopyMemory, which uses SEH to copy the memory (not in freeldr)
Make RtlLargeIntegerToChar use RtlpSafeCopyMemory to copy the string to the target buffer.
CORE-3767 #resolve
svn path=/trunk/; revision=57295
- Aric Stewart : While calculating lines, discard uniscribe data for non-visible lines.
- Nikolay Sivov : Release uniscribe data on Edit control destruction (valgrind).
- Sync Port wine.
- Someone needs to post a wine bug report due to the lack of support for Ctrl+Backspace. Debug fixme was added. See bug 1419.
svn path=/trunk/; revision=57291
- NdisReadConfiguration should return NdisParameterInteger for requests for NdisParameterHexInteger
- Assert that conversion to integer from string does not fail
- Always trigger the ISR function if MiniportInitialize or MiniportHalt is executing
CORE-6064 #comment Please try to replicate with DEBUG_ULTRA turned on in NDIS
svn path=/trunk/; revision=57289
Finally commit the patch to improve freeldr disk performance by Carlo Bramini.
Obviously someone wants to read 0 bytes and that should not fail.
svn path=/trunk/; revision=57283
* Make freeldr_common depend on bugcodes and remove this dependency from the other targets.
* Improve those conditions readability a bit.
svn path=/trunk/; revision=57280
Apply Carlos patch again, but use MaxSectors = 1 for now. That should result in the exact identical behaviour as before.
svn path=/trunk/; revision=57278
- Use a writable buffer for lpCommandLine in CreateProcessW. Simplify starting of control panel applets. Patch by Hermès Bélusca.
- CORE-6570 #resolve #comment We need FishEye ;)
svn path=/trunk/; revision=57277
- Remove the FILESYSBUFFER from the memory layout, grow the DISKREADBUFFER to 64 KB
- Make use of the full disk buffer when reading sectors, speeds up hdd boot process by about a factor of 2, patch by Carlo Bramini
- See CORE-6071
svn path=/trunk/; revision=57274
Stop using the FILESYSBUFFER. It's a 64 KB memory region statically set up int the lower 1MB address space. This memory is rare and it doesn't make sense to put the buffer there, since it's only used to copy the data from the DISKREADBUFFER (which *must* be in the lower 1MB) to a temporary location to copy it to some other location afterwards. It's replaced with temporary heap buffers and direct reads, where possible. Also Don't read a full FAT cluster, when only a part of it is required.
svn path=/trunk/; revision=57267
[NTOSKRNL]: As a side effect, implement SeCreateClientSecurityFromSubjectContext and SeGetTokenControlInformation, although most of the code was already there.
[NTOSKRNL]: One useful change, SE_GROUP_USE_FOR_DENY_ONLY (Deny Only Groups) are now handled correctly.
svn path=/trunk/; revision=57261
[PSDK]: Sync some headers and idl files with Wine.
[PSDK]: Import xmldom.h from Wine.
[LIBWINE]: Sync some debugging functions with Wine.
[UUID]: Add SID_SContainerDispatch.
[MSXML]: Import from Wine.
[MSXML2]: Import from Wine.
[MSXML3]: Sync with Wine 1.5.12.
[MSXML4]: Import from Wine.
[MSXML6]: Import from Wine.
[SYSSETUP]: Register the newly imported msxml* dlls.
svn path=/trunk/; revision=57257
- Move exception stuff to rtltypes.h
- Add missing exception flags
- Add _DISK_GEOMETRY_EX forward declaration
[PSDK]
- Add missing exception flags to winnt.h
[NDK]
- Remove exception flags, they are defined in both winnt.h and wdm.h
svn path=/trunk/; revision=57255
* Remove the importlibs folder. It's not needed anymore. While we're here, it's worth noting that the use of link_directories is not recommended, you should use target_link_libraries with absolute paths for example.
svn path=/trunk/; revision=57238
- Move more support for Desktop Proc into Win32k. Increase Desktop drawing by about 10%. Rearanged code and plugged in the server side callout.
svn path=/trunk/; revision=57236
[NTOSKRNL]: Restore the page table reference counting mechanism, and put it in a macro to be cleaner. Also use macros for testing PD boundaries, instead of math-by-hand.
svn path=/trunk/; revision=57229
Thank you thfabba (Thomas Fabber) for one of the most useful, yet ignored, tests in our suite. Hope this makes it up to you.
svn path=/trunk/; revision=57225
[NTOSKRNL]: Kill a bunch of now-useless DPRINT1s in Mm, significantly cleaning up the log.
[HIVES/MSI]: .NET needs InstallRoot, not InstallDir. Kills millions of DPRINT1s, thus cleaning up the log.
svn path=/trunk/; revision=57221
[NTOSKRNL]: No longer support non-ARM3 sections in NtProtectVirtualMemory, as the only OS calls were already NO-OPS.
[NTOSKRNL]: Always use ARM3 sections unless SEC_PHYSICAL_MEMORY is used, and make the check explicit.
[NTOSKRNL]: No longer support allocating memory on top of non-ARM3 sections.
[NTOSKRNL]: No longer ASSERT when certain features are not yet implemented, instead return an error code.
[NTOSKRNL]: Add another check in NtFreevirtualMemory when invalid memory is being freed, insert of ASSERTing.
[NTOSKRNL]: Implement and use MiIsEntireRangeCommitted when protecting memory to make sure the entire range is committed.
This patch removes multiple hacks, ASSERTs, and evil mixing of ARM3 and non-ARM3 code/memory.
svn path=/trunk/; revision=57215
Mark intrinsics on gcc with __attribute__((artificial)) so that they appear as one unit in the debug info. This way traces don't point inside the implementations, when they cause an eception, but rather to the line where they are used.
svn path=/trunk/; revision=57214
[NTOSKRNL]: Remove an ASSERT(FALSE) that was only there for testing.
[NTOSKRNL]: Support transition pages during prototype PTE faults, which is our first try at soft faults! Should fix ASSERTs that were seen in the previous attempts in ole32, corrupting the registry.
[NTOSKRNL]: It's fine for MiCreatePagingFileMap to fail in MmCreateSection -- don't assert and simply return failure. Should fix the ASSERTs taht were seen in KmTest.
[NTOSKRNL]: Enable richard's ARM3 section code unconditionally for all non-file backed sections. Works4me. Let's see what Testbot says.
Nobody has showed me how to use/where is PatchBot, and google founds 0 relevant results, so this is going into main again. However I'm actually home this week to revert if something goes wrong :)
svn path=/trunk/; revision=57209
Freeldr size is currently limited to 448 KB. On MSVC it was already at 442 KB, before wine's wctype table was used. The new wctype table is itself 37 KB. This lead to freeldr overflowing into memory regions that were used for the filesystem buffer, causing bootfailures. Fix this by giving freeldr it's own using _isctype(), since freeldr casts WCHAR to CHAR anyway.
svn path=/trunk/; revision=57205
Move imagesoft to applications and add it to build.
Convert lang files to UTF-8.
Fix casts and x64 warnings.
Patch by Hermes Belusca.
See bug 7185 for details.
svn path=/trunk/; revision=57201
Add vgafontedit to build.
Convert the language resource files into UTF-8.
Fix x64 warnings.
Patch by Hermes Belusca.
Bug 7152.
svn path=/trunk/; revision=57200
- Fix a memory overwrite bug in some crappy WINE code that leads to nasty memory corruption of the service cache entry. This corrupts the first field of the cache, causing it to always reallocate memory because it thinks the size is 0. This would be even worse if the code didn't overwrite the other 12 bytes of corruption with proper data on the next lines. This also caused the aliases value to be garbage because it pointed at uninitialized data.
svn path=/trunk/; revision=57199
- Implement proper support for partial sends on TCP sockets
- This prevents a deadlock when a very very large packet is queued to be sent
svn path=/trunk/; revision=57186
* Prefer STREQUAL over MATCHES since we're comparing with strings here. Brought to you by Hermès Bélusca.
See issue #7306 for more details.
svn path=/trunk/; revision=57181
- Add some tests for lstrncpyW
- This function is fine, I wrote that when chasing some bug thinking this function was at fault.
svn path=/trunk/; revision=57179
- Fix one of the worst bugs in AFD of all time. Our AFD code operated under the assumption that none of the input parameters would change once we called LockRequest. This assumption is completely false. The only guarantee that made was that the pages never disappeared from us, not they they couldn't be modified. There are frequent cases where the user-mode buffer was modified from underneath us (WSPRecv allocates the struct on stack which makes it invalid during overlapped operations that complete later). When this happened, we would bugcheck when we tried to unlock the buffers since we accessed this in a member of the struct the caller passed us.
- I've fixed this by adding a parameter to LockRequest which specifies whether the buffer should be copied back when it is unlocked.
- This bug has been around for ages and I was never able to figure out why we just freed garbage sometimes. Now that the ws2_32_winetest exposed it reliably, I was finally able to fix it.
svn path=/trunk/; revision=57175
- mark DllMain as a weak symbol for GCC.
- supply a stubbed DllMain for MSVC.
- DllMain is optional, and some DLLs don't implement it. That doesn't mean that they have no entry point, it means "I have nothing more to initialize than the CRT".
svn path=/trunk/; revision=57171
- Change AFD to a very aggressive buffering policy
- It now attempts to keep its 64K receive buffer completely full at all times
- This increases network performance significantly
svn path=/trunk/; revision=57170
- Implement LsaSetSeceret/LsarSetSecret (secret values are not encrypted yet) and LsaSetSystemAccessAccount.
- Improve some TRACE messages.
svn path=/trunk/; revision=57167
- Make hard-coded calling convention exceptions in ExInterlocked compatible with GCC 4.7, and add the corresponding TODO about how ugly this is
svn path=/trunk/; revision=57166
PSEH2 uses a special mechanism to deal with nested try blocks inside the same function. Instead of pushing a second exception registration record on the exception list, it uses an internal pointer to handle the different try-levels.
But when a function using SEH is inlined, the resulting code will push 2 registration frames on the stack. Now it happens with GCC 4.7.1 that these frames get shuffled on the stack, putting one for the inner try on a higher stack address. This is something that RtlUnwind regards as a bug and throws a STATUS_INVALID_UNWIND_TARGET exception. This was the reason for the crashing PSEH2_TEST, when compiled with GCC 4.7.1.
To fix this, I added a mechanism that will prevent functions using SEH from being inlined.
svn path=/trunk/; revision=57159
- Check result of walking the import descriptor and properly fail in case it wasn't successful.
See issue #7289 for more details.
svn path=/trunk/; revision=57156
[NTOSKRNL]: Remove an ASSERT(FALSE) that was only there for testing.
[NTOSKRNL]: Enable richard's ARM3 section code unconditionally for all non-file backed sections. Works4me. Let's see what Testbot says.
svn path=/trunk/; revision=57155
- Add some optimizations (GCC) on compilation of libMesa
- Move opengl32 to the opengl directory
- Disable ICD refcounting for now, it does more harm than good.
svn path=/trunk/; revision=57153
* Call cl directly instead of its full path using <CMAKE_C_COMPILER>. Figuring out why it didn't work is left as an exercise for the reader.
svn path=/trunk/; revision=57152
Hermès Bélusca:
* Fix a bug in the way we compiled the boot sectors on the x64 build.
* Consistently set the mc compiler, like gcc builds.
* Use proper variables when referring to the compilers.
See issue #7297 for more details.
svn path=/trunk/; revision=57151
- Fix a bug causing FIONREAD to return an incorrect value after a datagram receive completes when only returning a subset of the data (returns WSAEMSGSIZE)
svn path=/trunk/; revision=57147
- Add fake iostream headers to fix MSVC compilation
- hack numerical_limits<float>::infinity() implementation until we have a real C++ library
Oddly, empty headers seem to be enough for now
svn path=/trunk/; revision=57136
[IPHLPAPI] - Fix a use after free in getNthInterfaceEntity (CID 716795 - partly reverts r53190). Fix a possible double free in getArpTable (CID 716796).
svn path=/trunk/; revision=57127
- Use wine's wctype table in CRT for iswctype() function. Previous implementation was returning incorrect information for non-Latin characters. Problem found and fixed by Sergey Shamanaev (seven_ro).
- Move wctype.c from libwine to lib/sdk/crt instead of making yet another copy of it (currently there is one more in tools/unicode). Suggested by Amine Khaldi.
See issue #5090 for more details.
svn path=/trunk/; revision=57119
- Add support for getting and setting SO_EXCLUSIVEADDRUSE
- Add support for setting SO_REUSEADDR
[AFD]
- Implement proper TDI way of indicating a reusable address file
- More work remains in TCP/IP to support this
svn path=/trunk/; revision=57115
- Fix screensave access hangs but it was recoverable and did not crash. The new ATI points seem to kick in so disabling them for now and leaving error prints. See bug 7225, need to separate the message queue.
svn path=/trunk/; revision=57114
- Active, Focus and Foreground. Fixes many of the the wine tests and including the todos. Fix bug 7089.
- Get ATI support points in place, include code support for ptiSysLock which will be rewritten soon.
- Move Desktop Proc to Win32k. This will allow the access checks to work. Note: DesktopBG.c creates a Win32 thread every time a Desktop is created. Please keep the file and code intact for historical purposes.
- Get code close to wine.
References:
Focus test ROS passes: http://www.winehq.org/pipermail/wine-patches/2012-August/116887.html
Install acts and looks better: http://www.resmap.com/instructions/plug-ins/ERViewer7_0_RC.exe
svn path=/trunk/; revision=57111
- Attempt to fix Win test_mouse_input, properly set MA_ACTIVATE with right conditions when sending WM_LBUTTONDOWN.
- Get code close to wine.
svn path=/trunk/; revision=57109
Remove "size_t to DWORD/int..." conversion warnings in x64 build.
Patch by Hermes Belusca.
I replaced size_t by SIZE_T, as this is the proper Win32 type. And I also used the prefix 'cch' (count of characters) instead of the generic 'dw'.
See issue #7149 for more details.
svn path=/trunk/; revision=57108
- Move WM_SETREDRAW and WM_SYNCPAINT to Win32k.
- Patch by Alex Henrie: Minimize and maximize boxes can only appear when WS_SYSMENU is set.
svn path=/trunk/; revision=57105
- AttachThreadInput, this should fix the crash. Original crash patch by Giannis Adamopoulos. See bug 7225.
- WARNING: This code should be turned off until the restructuring is finished. Please read bug 7225!
svn path=/trunk/; revision=57103
- Fix a memory leak in the failure path of WSPGetSockName and WSPGetPeerName
- Fix a broken debug print in WSPGetPeerName
svn path=/trunk/; revision=57097
- Sync WIDL to Wine 1.5.10. Part 1 of 2.
- Adjust files to use MIDL compatible header guards.
- Add support for nameless structs and unions.
svn path=/trunk/; revision=57095
- Add more option support to WSPGetSockOpt and WSPSetSockOpt
- Return the correct error code for an unsupported option in WSPGetSockOpt
- Hack a SO_ERROR get handler together for apps that need it
- Add better unsupported protoopts debugging (particularly for getsockopt)
- SMPlayer can now stream media from the Internet :)
svn path=/trunk/; revision=57094
- Do not assume null termination of DllName and optimize extension search in LdrpLoadDll. Patch by Alexander Yastrebov (menone7 at gmail dot com).
See issue #7251 for more details.
svn path=/trunk/; revision=57092
- Fixed skip test regression in User32 Win wine test.
- Regressed one scroll control test. AbiWord scrolls and Explorer redraw issue still
the same.
- WIP: Rearranged code for standard structures prior to moving scroll code to Win32k. Proc support will remain in User32 (A2W etc) and redirected to Win32k. Drawing support will be added when this switch is made. Then ReactOS will have 100 percent wine scrolls plus our fixes in one file.
svn path=/trunk/; revision=57091
Patch by milawynsrealm:
* Implemented the beginnings of InitiateShutdownA/W (nothing works yet, and it's not exported).
* Running InitiateSystemShutdownA/W invokes InitiateSystemShutdownExA/W while
using "Other (Planned)" as the reason code.
- The "Other (Planned)" reason code seemed to be the best default
option so far.
Let me know if you feel otherwise.
* Added the shutdown flags that are used for InitiateShutdownA/W inside of
winreg.h.
* Include the reason.h header inside the winreg.h header so that the reason
codes can be
properly called. This conforms to the VC2010 method of adding this support.
* Added the declarations for InitiateSystemShutdownExA/W and InitiateShutdownA/W inside of winreg.h.
* In winuser.h, change the definitions for EWX_* to conform to MSDN
documentation
-
http://msdn.microsoft.com/en-us/library/windows/desktop/aa376868%28v=vs.85%29.aspx
* Added definition support for EWX_HYBRID_SHUTDOWN which is found in Windows 8.
* Calling InitiateSystemShutdownExA will convert strings to UNICODE and send it
over to InitiateSystemShutdownExW. The same is also for InitiateShutdownA.
See issue #7245 for more details.
svn path=/trunk/; revision=57089
- Deactivate Test_SimpleParameters in AttachThreadInput tests until it doesn't crash in win32k.
- Add some more tests in Test_SimpleParameters that should bring win32k to its limits
svn path=/trunk/; revision=57085
- change error messages in CallWindowProc to make it apparent that we caught an exception with SEH as requested by Caemyr
svn path=/trunk/; revision=57084
- Do not treat the release of the windows key as a hotkey if it was used as a modifier for another hotkey that was pressed
- Now if I press win+R only the Run window opens
svn path=/trunk/; revision=57083
- Hack around the fact that desktop handling in ros is awful. Now the child threads use the proper desktop.
- Convert all calls to SendMessage to SendMessageTimeout to make sure it won't hang (it doesn't hang in windows but who knows what can happen in ros)
- Add a few more tests
svn path=/trunk/; revision=57079
Use FILE_FLAG_NO_BUFFERING and FILE_FLAG_WRITE_THROUGH in the hope that they prevent journal changes from getting lost.
svn path=/trunk/; revision=57077
Use "advanced C++ features" like default parameters to reduce the complexity of StringOut calls and fix a warning in the MSVC build :-)
svn path=/trunk/; revision=57076
- Fix bug 6751, 7228 and others, Validate link windows, not relink them per say.
- Stop spurious window handle errors. This fixes all of the DeferWinPos handle errors.
svn path=/trunk/; revision=57068
- Increase the size of the ID buffers to fit certain non-standard IDs reported by Hyper-V on Windows 8. These should be allocated from pool, but this code really sucks and should be replaced anyway.
- Fixes buffer overflow causing a bug check during boot on Hyper-V
svn path=/trunk/; revision=57067
- Fix a critical bug in interrupt handling that could result in HAL returning the current processor to the wrong IRQL after an interrupt is handled that requires a raise to a synchronize IRQL.
svn path=/trunk/; revision=57065
- Fix a critical bug in interrupt sharing between scsiport miniports and other drivers which caused scsiport to mask the interrupts from other devices sitting behind it
- This commit and r57063 fixes bug 7231
svn path=/trunk/; revision=57064
- Fixed crash, see bug 7226.
- AttachThreadInput is new to ReactOS, it passes the wine tests, help is needed, white papers are welcome.
svn path=/trunk/; revision=57045
[NTOSKRNL]: Massively cleanup triplicated code that is just a series of macros/inlines.
[NTOSKRNL]: Handle more complex prototype PTE faults that are not just demand-zero.
In my private branch, I can now run ReactOS with *all* pagefile-backed sections running under ARM3 mode.
svn path=/trunk/; revision=57033
- Sync FormatMessage implementation to Wine 1.5.4. Fixes crash in kernel32_winetest:format_msg.
See issue #7217 for more details.
svn path=/trunk/; revision=57031
- Implement CaptureUnicodeStringOrAtom, which handles user mode class names or atoms, and use it in some places.
- Fix a number of format strings
svn path=/trunk/; revision=57023
* Set the attribute before the variable in write_func_param_struct(). Should be sent upstream.
* Initialize in an MSVC compatible way. Should be sent upstream.
[REACTOS/IDLS]
* Use widl instead of midl in MSVC builds.
svn path=/trunk/; revision=57019
- Use trivial Clip object, if NULL one is passed in IntEngAlphaBlend
- In NtGdiDescribePixelFormat allow passing NULL for the output pixel descriptor and copy the result *to* the caller not from the caller.
- Fix some format strings
- Update some annotations
svn path=/trunk/; revision=57013
Partly revert r56995. Firstly it breaks MSVC builds, secondly the stuff doesn't belong into this header. MS headers don't have __fpclassify etc and our headers are not supposed to contain any fancy additions. If you need it for 3rd party code, put it somewhere else.
svn path=/trunk/; revision=57012
[NTOSKRNL]: Set MiSessionSpaceWs address and use where needed.
[NTOSKRNL]: Initialize the session working set, and add support to MiLockWorkingSet and MiUnlockWorkingSet for session WS.
[NTOSKRNL]: Implement MiSessionAddProcess and MiSessionRemoveProcess which does the work required for this.
[NTOSKRNL]: Fix IoGetIrpRequestorProcess.
In my private branch, MmMapViewInSessionSpace is working fine for the user and GDI heaps. Will enable after TestBot tests.
svn path=/trunk/; revision=57011
- pass COLORADJUSTMENT to IntEngStretchBlt
- IntEngStretchBlt, EngStretchBltROP, DrvStretchBltROP always use a source, ASSERT that
- annotate pbrDefaultBrush and psurfDefaultBitmap as _Notnull_
- Don't use uninitialized psoInput in EngStretchBltROP
- Use trivial CLIPOBJ instead of NULL in IntEngStretchBlt
- Always pass a source surface to EngStretchBltROP
svn path=/trunk/; revision=57008
- Fix PREFAST warnings related to printf format strings, empty SEH handler, some possible NULL pointer dereferences, some unchecked return values
svn path=/trunk/; revision=57005
[NTOSKRNL]: Don't assume all prototype PTEs are cached, in fact none are in ReactOS at the moment! This resulted in the wrong hardware caching bits for the PEB/TEB and NLS Data which are using ARM3 section code.
[NTOSKRNL]: Also implement dirty page bit for prototype PTEs.
svn path=/trunk/; revision=57000
* Leverage the object library option provided by CMake 2.8.8+ which will speed up the build even further.
* These wrapping conditions will go away as soon as we ship the new BE.
svn path=/trunk/; revision=56997
- Add some more C99 math functions
- Fix definition of __forceinline and FORCEINLINE for C99
- Fix and add definition of _aligned_malloc and _aligned_free
- Fix definition of is(w)blank, as a C99 function, it is not in msvcrt.dll and has to be implemented in msvcrtex
All of this taken from the mingw-w64 project
svn path=/trunk/; revision=56995
- Partial update of math.h
Someone used to mingw-w64 headers system should look into this, I NEED those C99 functions!
svn path=/trunk/; revision=56989
- Implement MiProtectVirtualMemory for VAD based allocator (it's very similar to the already implemented MiSetProtectionOnSection function). However it still doesn't fully work because support in other functions is missing (failed assertions in MiFlushTbAndCapture).
See issue #7216 for more details.
svn path=/trunk/; revision=56986
- Align the stack on a 16-byte boundary for x86 like any sensible compiler would, anyway. Should fix weird GCC 4.7 behavior.
svn path=/trunk/; revision=56985
- Add _PRAGMA_WARNING_SUPPRESS() macro
- Fix definition of __analysis_noreturn for PREFAST runs
- Annotate RtlAssert with __analysis_noreturn everywhere
- Add proper annotations to all DbgPrint and DbgPrintEx definitions
- Suppress warnings for use of ProbeForRead outside of SEH
svn path=/trunk/; revision=56983
* Explicitly mark the functions as stdcall (NTAPI).. -mrtd changes the default calling convention, but name-decoration isn't affected by this. The -mrtd feature has its origin in some older linux-code-mode, and it's considered for removal in GCC 4.8.
* Fix some warnings.
svn path=/trunk/; revision=56973
* In the journey of running his GCC 4.7.1 compiled bootcd, two issues were revealed. Thomas kindly investigated them and fixed them, and as a result the bootcd boots to shell !
[CSRSRV]
* Fix the output parameter of ProcessLUIDDeviceMapsEnabled case in NtQueryInformationProcess() to be ULONG instead of BOOLEAN (as required).
* ASSERT that the function succeeds.
[SMSS]
* Move this condition up to the commented-out part (where it belongs).
svn path=/trunk/; revision=56972
- Take translation into account when inverting an XFORMOBJ
- Always use the matrices from the DCATTR instead of DCLEVEL
- Mark Device-to-World translation as invalid when the Window Origin is modified
Fixes 7088
See issue #7088 for more details.
svn path=/trunk/; revision=56970
- Fix _XSTATE_SAVE definition for ARM
[NDK]
- #if out NtCurrentTeb() inline function for ARM, it's implemented in ARMLIB
[CRT]
- #if out some intrinsics in msc/intrin.h that don't exist on ARM
- Add _crt_va_start, _crt_va_arg and _crt_va_end for ARM
svn path=/trunk/; revision=56964
- Separate some _CI* functions into their own files and export them from ntdll. Patch by Vincenzo Cotugno.
See issue #7085 for more details.
svn path=/trunk/; revision=56961
FF 3.6 and Thunderbird 3 is no more and thus is hosted on the same spot as the older FF versions are.
Fox Audio is updated to 0.9.1 including head rev mpg123 lib and hosted on our svn, too overwriting the old fox audio player.
Closing Bug #7194 and #7195, too by adding Total Commander and Click-N-Type Virtual Keyboard
svn path=/trunk/; revision=56956
- Do not leave critical section twice in Win32CsrReleaseConsole. Patch by Hermès Bélusca
See issue #6622 for more details.
svn path=/trunk/; revision=56955
Fix some problems with the lock implementation
- Add a 'generation' to the lock IRPs so we know if we've reprocessed them on any
specific turn when we unlock something else.
- Simplify coalescing of lock regions to just rebuilding the region by going through
the shared lock list and expanding a new region with any overlaps.
- Simplify coalescing new shared locks into larger regions in the same manner.
svn path=/trunk/; revision=56953
- Keep a reference count for FCBs to prevent race conditions when closing multiple handles to the same pipe concurrently
- Acquire the pipe list lock in NpfsQueryDirectory to guard against concurrent deletion of pipes
See issue #7205 for more details.
svn path=/trunk/; revision=56950
- Ensure that account names are not in use when a new alias, group or user is created. Return a proper status code if the account name is alread in use.
- Initialize logon/password related data for new user account objects.
- Add missing prototypes and user account control bits to ntsam.h
svn path=/trunk/; revision=56936
Add support for MSVC ARM
Configuration script works as usual from VS 11 ARM command prompt, host tools is configured for x86 and must be compiled from a VS x86 command prompt.
svn path=/trunk/; revision=56935
- Handle transferring buffer to user mode in WH_CALLWNDPROC(RET) hooks if lParam is a pointer.
This fixes the gallium3d opengl ICD, so newest VMWare opengl Driver should now work.
Please TEST!
svn path=/trunk/; revision=56931
[NTOSKRNL]: Cleanup the page fault handler to help with debugging, fix a few bugs and missing perf counters. There's a lot of cut corners that need some love, will be slowly adding more asserts/sanity code to make sure things are working right.
svn path=/trunk/; revision=56926
Check if the name of a new account (alias, group or user) is used before the account is created. This check is needed because account names must be unique.
svn path=/trunk/; revision=56917
* Leverage the object library option provided by CMake 2.8.8+ which will speed up the build even further.
* Compared strings using STREQUAL.
* These wrapping conditions will go away as soon as we ship the new BE.
svn path=/trunk/; revision=56909
- Fix a regression from r56863. New logic is: the requested data size is sizeof(ULONG), so if the result read is less than that - continue to the next slot. Thanks to SomeGuy for pointing out the problem.
See issue #7147 for more details.
svn path=/trunk/; revision=56907
[NTOSKRNL]: Implement MmMapViewOInSessionSpace, MmUnmapViewInSessionSpace. Win32k needs to use these to we can test them.
svn path=/trunk/; revision=56906
Don't set the ebp register when doing INTs. We only need the output after the int and since the REGS structure is usually not fully initialized before calling Int386, we would load random values. This can cause crashes on VPC.
Thanks to hbelusca fo testing.
svn path=/trunk/; revision=56903
Implement check for SERVICE_WIN32_OWN_PROCESS or SERVICE_WIN32_SHARE_PROCESS type.
Patch by Hermes Belusca.
See issue #7179 for more details.
svn path=/trunk/; revision=56900
- When no file name is passed on the command line, open a file-open-dialog
- Halfplement printing
- Replace Quit button with install button
- Patch by milawynsrealm <spaceseel at gmail got com>
See issue #6803 for more details.
svn path=/trunk/; revision=56898
- Don't modify a dialog template resource directly, instead make a copy of it.
- Should be sent to wine (bug #7189)
[BROWSEUI]
- Remove useless diff file.
svn path=/trunk/; revision=56897
- Fix possible buffer overflow: ExpandEnvironmentStrings wants character count size of a buffer, not byte sized. Spotted by Thomas Faber.
See issue #6053 for more details.
svn path=/trunk/; revision=56888
- Revert r56865, it duplicates same action (expanding the var), spotted by Thomas Faber.
See issue #6053 for more details.
svn path=/trunk/; revision=56887
- Fix uninitialized variable usage warning, based on a patch by Edijs Kolesnikovics.
See issue #6828 for more details.
svn path=/trunk/; revision=56886
- Hermes Belusca: Fix wrong definition of CMHIVE structure (defined only 3 FileHandles when 4 is needed).
See issue #7162 for more details.
svn path=/trunk/; revision=56884
* Improve the way DDKAPI, DDKCDECLAPI and DDKFASTAPI are defined. With that we don't need ReactOS specific defines anymore.
* Already sent upstream.
svn path=/trunk/; revision=56882
- Use COUNT_OF where appropriate. Miscellaneous code cleanup. Patch by Hermès Bélusca
See issue #7182 for more details.
svn path=/trunk/; revision=56876
* Fix Cookie for loader locking generation
Silences "LdrUnlockLoaderLock() called with an invalid cookie!" debug messages
svn path=/trunk/; revision=56874
* Track pixel format with window handle for device contexts
* Fix SetPixelFormat in case pixel format were already set
svn path=/trunk/; revision=56873
- Update Uniata to 41b5. Thanks to Alter for the great driver and Olaf Siejka for preparing the patch, getting our changes upstream and testing the updated version.
See issue #7148 for more details.
svn path=/trunk/; revision=56870
- Implement Import/Export of hive files. Patch by Hermes Belusca (German translation by me)
See issue #7180 for more details.
svn path=/trunk/; revision=56869
- Apply the "Refresh" command to the ListView in addition to the TreeView
- Use COUNT_OF where applicable; miscellaneous readability and consistency fixes
- Space fixes in resource files; French translation fixes
- Patch by Hermes Belusca
See issue #7180 for more details.
svn path=/trunk/; revision=56868
- Edijs Kolesnikovics: Expand environment strings during processing startup items in HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Run.
See issue #6053 for more details.
svn path=/trunk/; revision=56865
- Don't stop enumerating functions of a PCI device if current function returned invalid vendor id. Also add a check for returned data size. This fixes detection of some storage controllers, e.g. Intel ICH8, ICH9, ICH10 chipsets.
Thanks to Alter (UniATA author) for finding the problem and providing solution.
See issue #7147 for more details.
svn path=/trunk/; revision=56863
Correct comments and code error introduced in rev 56844.
Patch by Hermes Belusca.
I used a helper variable to determine whether or not ScmDeleteNamedPipeCriticalSection must be called. Changing the order in which functions are called could cause a deadlock.
See issue #7172 for more details.
svn path=/trunk/; revision=56861
Allocate the ARC_DISK_SIGNATURE structures from the heap instead of putting them into the initial loader block allocation, since the number of disks changes later. Fixes memory corrruption and boot breakage when 3 hdds are connected.
[NTOSKRNL]
Remove a hack that disabled unloading drivers, which didn't work because of memory corruption from freeldr. This works now, too.
See issue #7139 for more details.
svn path=/trunk/; revision=56860
- Make sure freeldr is not compiled with SSE instructions (default for VS 11+)
- Print trap number and instruction stream on bugcheck
- Fix assembly code for ML
svn path=/trunk/; revision=56857
- Implement SampGetObjectAttributeString to retrieve the size of a string, allocate a buffer and retrieve the string from the registry.
- Use SampGetObjectAttributeString to retrieve string attributes in SamrQueryInformationXxx functions.
- Add most missing cases to SamrQueryInformationUser.
- Fix a type mismatch in ntsam.h.
svn path=/trunk/; revision=56850
- Wine doesn't prepend \registry\ to any registry path, which is wrong. Fix that.
- Bring back previous implementation of GetGeoInfo & co.
- Fix GEO_OFFICIALLANGUAGES in the headers.
- Thanks to Kamil Hornicek and Samuel Serapion for finding the problem and creating patches.
See issue #6595 for more details.
svn path=/trunk/; revision=56849
Implement service database locking via RLockServiceDatabase / RUnlockServiceDatabase etc... and use this functionality.
Patch by Hermès Bélusca.
I did not rename the database lock functions because the functions names RLockServiceDatabase and RUnlockServiceDatabase are misleading. They do not lock any database but lock the service start function. The existing functions ScmLockDatabaseExclusive, ScmLockDatabaseShared and ScmUnlockDatabase do protect (lock) the service database (service, group and image lists) form concurrent access. That is also the reason for renaming the SC_LOCK_STRUCT to START_LOCK.
See issue #7131 for more details.
svn path=/trunk/; revision=56844
- Simplify AcquireRemoveRestorePrivilege. Improve clean-up & acquire privilege before unloading hive in CreateUserProfileW. Patch by Hermes Belusca
See issue #7160 for more details.
svn path=/trunk/; revision=56843
- Improve permission acquirement for loading/unloading hives. Patch by Hermes Belusca
See issue #7159 for more details.
svn path=/trunk/; revision=56835
- Acquire the necessary privileges when loading/unloading hives. Patch by Hermes Belusca.
See issue #7159 for more details.
svn path=/trunk/; revision=56825
- Store all fixed size user attributes in the registry value "F".
- Initialize fixed size domain attributes in SampCreateUserAccount and SamrCreateUserInDomain.
- Implement most information classes of SamrQueryInformationUser.
- Disable all informationclasses of SamrSetInformationUser that store fixed size attributes.
svn path=/trunk/; revision=56820
- Implement missing cases of SamrSetInformationDomain.
- Use the fixed domain data to assign new RIDs in SamrCreateAliasInDomain and SamrCreateUserInDomain.
svn path=/trunk/; revision=56819
- Implement all cases of SamrQueryInformationDomain.
- Store all fixed size domain attributes in the registry value "F".
- Initialize all domain attribute in SampCreateDomain.
svn path=/trunk/; revision=56818
- Implement SURFACE_vSetPalette and use it everywhere to set the palette for a surface. Make SURFACE::ppal const, so noone accidentally modifies it without properly handling the references.
svn path=/trunk/; revision=56816
- Use a define for the service name in QueryServiceConfig2 test. Patch by Hermes Belusca. Bug 7145.
- Add LockDatabase test for Lock/UnlockServiceDatabase, QueryServiceLockStatus. Patch by Hermes Belusca. Bug 7146.
svn path=/trunk/; revision=56810
- Add test for QueryServiceConfig2 with SERVICE_CONFIG_DESCRIPTION and SERVICE_CONFIG_FAILURE_ACTIONS. Patch by Hermes Belusca.
- Remove ok_lasterr in favor of ok_err.
See issue #7144 for more details.
svn path=/trunk/; revision=56809
Handle read only devices and report them unsupported.
Forbid the formatting of the system drive.
Patch by Mikael Lyngvig.
See bug 2081 for more details.
svn path=/trunk/; revision=56808
Implement RtlConvertVariantToProperty, RtlConvertPropertyToVariant, PropertyLengthAsVariant
Those are just wrappers around Ole32 functions.
svn path=/trunk/; revision=56795
- Really implement RtlRandomEx and add its values table
- Rename SavedValue to RtlpRandomConstantVector to allow RtlRandom & RtlRandomEx to have two different values tables
- Complete and fix documentation
- Stop using magic values
svn path=/trunk/; revision=56791
- Initialize Winsock before using getaddrinfo (and in general). Fixes test regressions from r56773.
See issue #7135 for more details.
svn path=/trunk/; revision=56789
- Implement SamGetAliasMembership, SamQueryInformationAlias and SamSetInformationAlias.
- Add stub SamLookupNamesInDomain.
[SAMSRV]
- Implement SamrGetAliasMembership and SamrSetInformationAlias.
- Add stub for SamrQueryInformationAlias.
- SamrCreateAliasInDomain and SAM setup code: Set the description attribute for aliases.
svn path=/trunk/; revision=56785
Properly export RtlVolumeDeviceToDosName by using IoVolumeDeviceToDosName. No need to reimplement RtlVolumeDeviceToDosName.
svn path=/trunk/; revision=56781
[XINPUT1_3] - Move header file to PSDK, where it belongs
[MLANG][OBJSEL] - Remove unneeded import library references
[README.WINE] - Mark modules as synced to 1.5.4 if they are
svn path=/trunk/; revision=56766
Code cleaning and removing bugs.
Patch by Hermès Bélusca.
I did not include the changes to EnumServicesStatusW because it does not improve the code but introduces an unnecessary exception.
See issue #7127 for more details.
svn path=/trunk/; revision=56763
- Sync to Wine 1.5.4. Resource files left untouched.
- Fix INTERNET_OPTION_SECURITY_CERTIFICATE_STRUCT if SSL support unavailable (already applied upstream)
See issue #7117 for more details.
svn path=/trunk/; revision=56760
- Better mark ROS-specific changes
- Make wininet_ros.diff sensible
- Update wininet.h
See issue #7117 for more details.
svn path=/trunk/; revision=56758
- Do not manually allocate an incompatible hostent structure in FindEntryInHosts. Use populate_hostent instead. Fixes buffer overflow in name resolution.
See issue #7126 for more details.
svn path=/trunk/; revision=56756
intsafe.h:
- Fix type redefinitions when ntdef.h is included
- Implement (Rtl)LongLongSub and (Rtl)UlongLongMult
- Add some more definitions
- misc fixes
svn path=/trunk/; revision=56750
- Fix wait asserts to account for the Thread->WaitNext case
- At least output a warning in place of the assertion failure in KeWaitForMultipleObjects
See issue #7129 for more details.
svn path=/trunk/; revision=56749
- Disable the assert in KeWaitForMultipleObjects for now, as tcpip isn't straightforward to fix
See issue #7129 for more details.
svn path=/trunk/; revision=56748
Improve a check in RQueryServiceConfig2W and RQueryServiceConfig2A
Patch by Hermès Bélusca.
See issue #7123 for more details.
svn path=/trunk/; revision=56739
Implement SERVICE_CONFIG_FAILURE_ACTIONS case in RChangeServiceConfig2W.
Patch by Hermès Bélusca.
I moved the changes to RChangeServiceConfig2W into a new function (ScmSetFailureActions). I also removed the check for services that run in the services.exe process because ReactOS will not support this kind of services. We should use a separate process or svchost instead.
See issue #7122 for more details.
svn path=/trunk/; revision=56738
* Alter the way we handle the dlldata generation with midl. This works around the warnings we get about multiple rules to generate proxy.dlldata.c.
* Brought to yo by Peter Kummel.
svn path=/trunk/; revision=56731
When a VAD is truncated using VirtualFree, truncate the corresponding MemoryArea
so new allocations will match in both spaces.
svn path=/trunk/; revision=56727
This new patch corrects bugs introduced in the implementation of
SERVICE_CONFIG_FAILURE_ACTIONS case in RQueryServiceConfig2W() and
RQueryServiceConfig2A().
Patch by Hermès Bélusca.
See issue #7101 for more details.
svn path=/trunk/; revision=56725
Removing the redundant function WaitForLSA() and use ScmWaitForLsa() instead.
Patch by Hermès Bélusca.
See issue #7110 for more details.
svn path=/trunk/; revision=56721
Implement SamEnumerateAliasesInDomain and SamrEnumerateAliasesInDomain. SamEnumerateAliasesInDomain does not return the names of the enumerated aliases, most likely due to a bug in rpcrt4.dll.
svn path=/trunk/; revision=56718
- Use IRegistrar mechanism for registration. Fixes urlmon:misc and mshtml:dom crashes
See issue #7113 for more details.
svn path=/trunk/; revision=56717
Fix first argument of the target program so that paths with spaces aren't
reparsed by the c runtime as multiple arguments.
svn path=/trunk/; revision=56716
- Fix attach thread input based on which thread is foreground. Tested with wine Msg test_SetFocus, CursorIcon test_SetCursor and test_ShowCursor, still needs more testing. See bug 7098 and bug 7107, used the test cases and it works with Qemu.
svn path=/trunk/; revision=56714
- Add missing C definition of InlineIsEqualGUID in guiddef.h
- Use portable TYPE_ALIGNMENT instead of __builtin_alignof in mstcpip.h
- Remove definition of DL_EUI48 from netiodef.h, since it's already definied as a forward declaration in mstcpip.h
- Make WS2TCPIP_INLINE extern inline instead of static inline to fix warnings about usage in non static inline functions
svn path=/trunk/; revision=56710
- Implement SamGetMembersInAlias and SamrGetMembersInAlias.
- Add registry helper routines SampRegQueryKeyInfo, SampRegEnumerateValue and SampRegSetValue.
- Remove unused SAM database code.
- Store the name and members key (only for aliases and groups) in the database object.
svn path=/trunk/; revision=56704
- Fast fix attach thread input. Tested with win Msg test_SetFocus, fixed the WM_ACTIVATE issue, still needs more testing. See bug 7098, used the test case and it works.
svn path=/trunk/; revision=56703
- brasilian portugeese translation by Matheus Macabu (bug no. 7079);
[AHK]
- add scripts for including AHK tests into Reactos\Bin;
svn path=/trunk/; revision=56697
Implement SERVICE_CONFIG_FAILURE_ACTIONS case in RQueryServiceConfig2W and RQueryServiceConfig2A.
Patch by Hermès Bélusca.
See issue #7101 for more details.
svn path=/trunk/; revision=56696
- Implement SamSetInformationDomain.DomainNameInformation and SamrSetInformationDomain.DomainNameInformation.
- Set the account domain name for SAM too.
- Remove the old samlib.h file.
svn path=/trunk/; revision=56687
- Implement SamFreeMemory and SamLookupDomainInSamServer.
- Remove unused file debug.h.
[SAMSRV]
- Implement SamrLookupDomainInSamServer.
- Add registry helper functions.
- Store the domain name in a new Name attribute which is used by Sam(r)LookupDomainInSamServer.
svn path=/trunk/; revision=56685
- Add __post_invalid to specstrings.h
- Add missing InterlockedCompareExchangePointerAcquire and InterlockedCompareExchangePointerRelease for x86 to winbase.h
- Add missing MemoryBarrier definition
svn path=/trunk/; revision=56683
- Allow placement of a wait entry in virgin address space prior to mapping a page.
- Add a wait entry in the address space when paging in section view.
- Correct a problem that previously prevented us from evicting or querying pages that
had been protected into PAGE_NOACCESS protection by adding a query for disabled
mappings and handling disabled mappings in a few places (this may be controversial).
svn path=/trunk/; revision=56682
- Get the logon user SID using the new SAM keys and values. This is still a hack.
[SAMLIB]
- Remove all hacks.
[SYSSETUP]
- Remove all administrator account hacks.
svn path=/trunk/; revision=56680
- Implement SamOpenUser and SamSetInformationUser.
- Add SamQueryInformationUser stub.
[SAMSRV]
- Implement SamrOpenUser.
- Add a hack to SamrSetInformationUser that enables us to store a users password (as plain-text) in the registry.
- SamrCreateUserInDomain: Fix the creation of user account key names and start counting user RIDs at 1000.
- Clean-up the SAM registry setup code.
- Assingn SIDs to the BUILTIN and Account domains.
- Create the Administrator and Guest accounts in the SAM setup.
svn path=/trunk/; revision=56679
Move the creation of the random account domain SID from syssetup.dll to lsasrv.dll. This change is required because the account domain SID must be stored in the LSA database before the SAM database initializes. Syssetup.dll created the account domain SID much too late.
svn path=/trunk/; revision=56678
- Don't check for send nonclient paint messages and bump up the paint count. This may spell the end of the PAINTING BUG spam.
svn path=/trunk/; revision=56673
- Implement SamCreateUserInDomain and SamrCreateUserInDomain.
- SamOpenDomain: Use the correct object types when trying to open a domain object.
svn path=/trunk/; revision=56672
Add Names sub keys to the Alias, Groups and Users keys of the Account an Builtin domains. These will later contain the 'Name to SID' entries that will simplify lookup by name operations.
svn path=/trunk/; revision=56670
- Return the actual number of bytes copied for configuration descriptor requests
- Fix MSVC warnings and NULL vs 0 vs FALSE issues
svn path=/trunk/; revision=56667
- Add SxS for classes, still not plugged in for a year now.
- Need more information on global headers to support class contexts. ATM the headers are local, for noting what is needed.
svn path=/trunk/; revision=56652
[KERNEL32]: Simplify QueueWorkItem since the trampoline is in Rtl.
[KERNEL32]: Add SxS support to QueueUserApc.
[KERNEL32]: Add some extra sanity checks/assertions in certain places.
svn path=/trunk/; revision=56651
- Use RtlAnsiStringToUnicodeString instead of mbstowcs to correctly handle manifests that aren't null-terminated
See issue #6743 for more details.
svn path=/trunk/; revision=56649
- Use Wine debug messages in samlib.dll.
- Move SAM initialization code from samlib.dll to samsrv.dll.
- Lsass.exe initializes SAM instead of syssetup.dll.
svn path=/trunk/; revision=56648
- Convert to Wine debug messages.
- Add server side stubs of the SAM RPC interface. The RPC server is not implemented yet.
svn path=/trunk/; revision=56645
- Swap GetFullPathName success/failure cases in BasePushProcessParameters. Fixes kernel32:GetModuleFileName test.
See issue #6786 for more details.
svn path=/trunk/; revision=56643
- Partial sync port Wine 1.5.4 for system.c.
- Researched branches/GSoC_2011/ThemesSupport and current changes before sync.
svn path=/trunk/; revision=56642
- Add new ntlsa.h file to the PSDK.
- LsarAddPrivilegesToAccount: Implement the ability to add new privileges to an existing privilege set.
svn path=/trunk/; revision=56594
- Fix some sickening formatting disasters
- Allocate the unicode string buffers from paged pool, like everything else in the NDIS config API
svn path=/trunk/; revision=56590
- Handle unknown color types in TranslateCOLORREF
- properly translate the color back to RGB in NtGdiSetPixel
Fixes a few hundred thousand gdi32 winetests (dib) :)
svn path=/trunk/; revision=56582
- In DC_vSetOwner, do not take ownership of the current dc brush and pen, instead dereference them and select default brush and pen. Fixes bug #7070
- Fix debug logging in GDIOBJ_vDereferenceObject
- create debug log event in GDIOBJ_vSetObjectOwner
svn path=/trunk/; revision=56579
Do not add the "SeUnsolicitedInputPrivilege" privilege to the administrators privilege set because it has been superseded by the "SeMachineAccountPrivilege" privilege and the later has already been added to the privilege set.
svn path=/trunk/; revision=56578
- Fix bug 6796, use wine code to clear bits before checking to minimize or maximize.
- Fix all the wine Win test_window_styles and todo tests.
svn path=/trunk/; revision=56570
* Sync to Wine 1.5.4. I excluded loader and virtual tests from the sync until bug #7071 is fixed.
See issue #7070 for more details.
svn path=/trunk/; revision=56567
- Patch by Dmitry Timoshkov : SetParent() should use ShowWindow() to make a reparented window visible and move a window to new position.
- Note: SetWindowPos should be correct, http://www.winehq.org/pipermail/wine-cvs/2012-April/086267.html
svn path=/trunk/; revision=56560
- Fix the remaining wine Win tests for FindWindow/Ex/A/W. Check the buffer not the length, buffer could be pointing to a null. Enable the message window handling. Use wine code for FindWindowExA.
svn path=/trunk/; revision=56557
Add a 4th instance of fusion.dll back into system32.
(doesnt match any actual windows system but needed for wine's msi.)
svn path=/trunk/; revision=56556
- Make EngSecureMem probe the buffer, so we have at least a minimum functionality, even though it's not safe
- Implement EngSecureMemForRead
- Use IntEngBitBlt from IntEngStretchBlt, when source and dest size are equal
svn path=/trunk/; revision=56553
- Make CreateDIBPalette handle BITMAPCOREINFO (we need to handle this, since win32k can load bitmap files, that can contain these headers)
- Remove obsolete IntSetDIBColorTable, IntGetDIBColorTable
svn path=/trunk/; revision=56548
- Alloca to use GDIOBJ_vReferenceObjectByPointer with exclusively locked objects
- Make sure the global mono palette actually has 2 entries
- Simplify allocation of default palette
- Implement GreGetSetColorTable, replacing IntSetDIBColorTable / IntGetDIBColorTable
- Make sure that memory possibly copied to user mode is zeroed.
svn path=/trunk/; revision=56546
Fix a couple of omissions from my PAGEOP removal;
Make sure to place a wait entry in the address space when faulting
in bss pages.
Restore the original entry in the final cases of section swap out,
because the entry had been replaced with a wait and nothing else
restores it.
svn path=/trunk/; revision=56540
- Fix crash when sending activate application message and check if thread is in cleanup. Use kernel functions for thread data types.
svn path=/trunk/; revision=56538
- Implement LsaEnumeratePrivilegesOfAccount and LsarEnumeratePrivilegesAccount.
- Implement parts of LsarAddPrivilegesToAccount.
svn path=/trunk/; revision=56527
Do not ASSERT that the object is not exclusively locked in GDIOBJ_vDereferenceObject. The idea behind this was to easily detect cases where someone would accidentally dereference an object, instead of unlocking it. But this function is used from other functions, that can definately deal with exclusively locked objects. Changing this would leat to code duplication / more complex code.
Forgetting to unlock an object will still be detected by the kernel, when returning to user mode, since APC would still be disabled.
Should fix failed assertion when running dx9 setup.
svn path=/trunk/; revision=56525
- Create default hatch brush bitmaps on win32k init
- After creating a PDEV fill the fields in the ahsurf array that are not filled by the driver with these bitmaps
- don't create a new bitmap when a hatch brush is created, instead only store the style and use the related bitmap from the PDEV when realizing the brush.
- Fix the palette of DIB brushes, when realizing the brush
- Don't leak the bitmaps for DIB brushes in BRUSH_Cleanup
svn path=/trunk/; revision=56523
Implement CreateDIBPalette(), replacing BuildDIBPalette.
It creates a palette without a handle and handles DIB_PAL_COLORS as well as DIB_RGB_COLORS. Additionally it now handles DIB_PAL_BRUSHHACK, which is used for DIB brushes, when DIB_PAL_COLORS is passed to CreateDIBPatternBrush(). The palette is not created directly, but instead the indices into the DC palette are put into the palette and the real palette is created, when the brush is realized. The latter is not yet handled.
svn path=/trunk/; revision=56518
[NTOS]: Implement Case C of NtFreeVirtualMemory, which is sometimes seen in some heap logs. The hard part is figuring out the right amount of committed/decommitted pages. Only supports 2-level paging for now as the algorithm is already messy enough.
svn path=/trunk/; revision=56515
Rename PALETTE_AllocPalette2 to PALETTE_AllocPalette, allocate PALETTE and color array in one allocation, improve code
svn path=/trunk/; revision=56514
- Do not multiply define WINVER and friends
- Add missing _WIN32_WINNT define in rpcrt4_winetest. Already sent to Wine(TM)
svn path=/trunk/; revision=56513
- Fix warnings - there is no DischargePolicy[4]
- Disable tests that fail on Windows 2003 (576 failures?!)
- Add missing file header
svn path=/trunk/; revision=56506
- Add tests for CreateDIBPatternBrush(Pt)
- Add some helper code to initialize a dibsection and a palette, which can be reused by other tests
- Add tests for DIB brush in GetObject
svn path=/trunk/; revision=56498
- Add cjMaxBits parameter to GreCreateDIBitmapInternal
- Rename GDIBRUSH_* constants to BR_*, as they are named in gdikdx
svn path=/trunk/; revision=56496
- Fix SetCurrentDirectoryA to return success when setting the current directory again
See issue #6874 for more details.
svn path=/trunk/; revision=56490
Account for wait entries delaying a view attribute change and interacting
with the machinery in CoW copying code, specifically making CoW insert and
revoke a wait while copying the page, and make the CoW copy function take
both pages as arguments rather than relying on the address space.
Also alter the MM_WAIT_ENTRY value a bit to take into account an assert
that it must have bit 0x800 set.
svn path=/trunk/; revision=56488
- Add the linking exception text to the official ReactOS licensing text. Previously, the exception text was only available at the ReactOS website with no references from the source code.
svn path=/trunk/; revision=56479
A pointer is not a handle. A handle is not a pointer. At least in most cases. Defining STRICT in win32k revealed horrible things...
svn path=/trunk/; revision=56478
- Fix a palette leak in DIB_CreateDIBSection
- Remove PALETTE_AllocPalette and PALETTE_AllocPaletteIndexedRGB, use PALETTE_AllocPalWithHandle instead
svn path=/trunk/; revision=56476
* Implement Ctrl+C support on MessageBox. Patch by Guido Pola (prodito {at} live [dot] com).
See issue #5281 for more details.
svn path=/trunk/; revision=56475
Modify DIB_MapPaletteColors, first allocating a palette without initializing the colors, then setting up the colors. This wway we don't need to allocate an intermediate buffer.
svn path=/trunk/; revision=56474
- Implement GreCreatePaletteInternal and make NtGdiCreatePaletteInternal safe regarding user mode memory access.
- Allow allocating an indexed palette without initializing the colors
- Implement PALETTE_AllocPalWithHandle
- remove pointless Self member in PALETTE structure
- cleanup palette.h a bit
svn path=/trunk/; revision=56473
- Import isinf and isnan implementations from libwine_port and adjust port.h and config.h to support them
- Remove isinf/isnan hacks from jscript and wined3d build files
- Add missing port.h include to jscript/array.c (sent and applied upstream)
See issue #7027 for more details.
svn path=/trunk/; revision=56470
Rewrite SURFACE_AllocSurface(), which now does the part of the former SURFACE_bSetBitmapBits() as well. This is to allocate the bits for a bitmap directly with the SURFACE object in one allocation. Also don't use kernel mode sections anymore by default, but paged pool memory.
svn path=/trunk/; revision=56460
Don't calll GDIOBJ_vDereferenceObject from GDIOBJ_vUnlockObject, instead manually decrement the reference count. This allows us to ASSERT that the former function is never called with an exclusive lock held.
svn path=/trunk/; revision=56458
Never ever trust the wine file layout.
wine can load a dll when there is no file using builtins, while normal windows dont.
This fixes fusion beeing installed in the wrong places.
Loading the 2nd and 3rd copy of fusion fails atm.
svn path=/trunk/; revision=56453
Autogenerate oldnames for MSVC as well.
Currently on gcc builds we use a special def file syntax and use dlltool to create the lib, while on MSVC builds we use an asm file with aliases.
svn path=/trunk/; revision=56448
Fix a serious bug in PSEH that could lead to stack corruption and was causing CSRSS to crash, when large amounts of text were output on the console.
Background: PSEH used __builtin_alloca to allocate an SEH registration frame on the stack on demand, ie only for the first try level. But there are some nasty things with __builtin_alloca: First it DOES NOT - as one might think - free the allocated memory, once the allocation "goes out of scope", like with local variables, but only on function exit. Therefore it cannot normally be used inside a loop. The trick that PSEH used to "fix" this problem, was to save the stack pointer and reset it back at the end. This is quite problematic, since the rest of the code might assume that the stack pointer is still where it was left off after the allocation. The other thing is that __builtin_alloca() can allocate the memory whenever it likes to. It can allocate everything on function entry or not allocate anything at all, when other stack variables that went out of scope have left enough space to be reused. In csrss it now happened that the allocation was done before the stack pointer was saved, so the memory could not be freed by restoring the old stack pointer value. That lead to slowly eating up the stack since the code was inside a loop.
The allocation is now replaced with a variable sized array. The variable stays in scope until the _SEH2_END and will be automaticall cleaned up by the compiler. This also makes saving and restoring the stack pointer obsolete.
Reliability++
svn path=/trunk/; revision=56442
Remove PAGEOPs and use the argument to pass relevant information where needed,
including generally a section page table entry or a PFN_NUMBER.
In its place, the code looks for an MM_WAIT_ENTRY in the appropriate place and
waits, returning STATUS_MM_RESTART_OPERATION. Rather than copying the pagein
and CoW operations in the PAGEOP contention case, we just punt back and restart
the operation, resulting in simpler code.
svn path=/trunk/; revision=56441
When a DC has the default palette selected, use the color value from PALETTERGB() directly instead of looking it up in the dc palette (just as the code was before). Fixes skipped bitmap winetests
svn path=/trunk/; revision=56440
- In NtGdiSetPixel, return the real RGB value, which is the target format color translated back to RGB, not the desired RGB value. Fixes a few bitmap winetest.
- Handle a too large index in PALETTE_ulGetRGBColorFromIndex
svn path=/trunk/; revision=56439
- Do not null terminate buffer before initializing its length in RtlDosSearchPath_Ustr. Use unicode string functions made for this purpose instead.
See issue #7010 for more details.
svn path=/trunk/; revision=56438
- Add a small RtlException test, mainly to check that a stack overflow really throws STATUS_STACK_OVERFLOW, not STATUS_ACCESS_VIOLATION.
svn path=/trunk/; revision=56434
- add missing DBG_DECREASE_LOCK_COUNT() when deleting an object that was exclusively locked. Fixes strange hang in comctl32_winetest.
The question is: why does changing a variable, which is never checked cause such trouble?
- Don't pass the DC's clip object to IntEngCopyBits, when querying a pixel
- make acExclusiveLockCount[] 0x20 bytes long, just to be really sure, that (x & 0x1F) does never overflow it.
svn path=/trunk/; revision=56433
- Do not rely on undefined behavior to get values easily accessible as constants.
- Quick build fix. Will be sent to Wine, possibly modified.
svn path=/trunk/; revision=56432
- Implement TranslateCOLORREF, based on IntHandleSpecialColorType, but now returning both the translated color and the resulting RGB color, correctly handling DIBINDEX, which will use the given value without translation.
- Rewrite NtGdiSetPixel using IntPatBlt
- Rewrite NtGdiGetPixel using IntEngCopyBits
- Fix EXLATEOBJ_iXlateTable to return 0, when the palette index is too big, instead of wrapping around
- Remove old debugcode
svn path=/trunk/; revision=56429
Pass EBRUSHOBJ to IntPatBlt instead of the BRUSH. This way we can use the DC's eboFill, instead of initializing a new EBRUSHOBJ.
svn path=/trunk/; revision=56427
- Accept NULL Buffer/Size in RtlGetFullPathName_Ustr
- Fix NULL dereference in RtlGetFullPathName_UstrEx if no StaticString was given
- Various other corrections to path functions. 17 test failures fixed overall.
svn path=/trunk/; revision=56426
- Make XFORMOBJ_iSetXform fail, when the XFORM is invalid (eM11 or eM22 are 0)
- Handle error in GreModifyWorldTransform
svn path=/trunk/; revision=56422
- Add a test for RtlDoesFileExists_U (and its three friends, if addresses are provided)
- Various small path test improvements
svn path=/trunk/; revision=56416
- Add a test for RtlGetFullPathName_Ustr (requires the function address to be known)
- Various improvements to the other path tests
svn path=/trunk/; revision=56413
- Add test for RtlGetLongestNtPathLength
- Add test for RtlDetermineDosPathNameType_U (and RtlDetermineDosPathNameType_Ustr, if address is known)
svn path=/trunk/; revision=56411
- Add missing mmtypes.h include in mmfuncs.h
- Remove RtlDetermineDosPathNameType_Ustr from rtlfuncs.h, move it up in path.c
- Add RtlGetLongestNtPathLength to rtlfuncs.h
svn path=/trunk/; revision=56410
- Fix a bug in the region code that could cause allocating 0 bytes
- Remove SetLastError()
- "flatten" IntGdiCombineRgn making it more readable
svn path=/trunk/; revision=56409
- Add forgotten storage of the LUN in the PDO extension which not only broke units beyond LUN 0 but also caused a BSOD due to reporting duplicate devices to the PnP manager
- Thanks to tower for reporting
svn path=/trunk/; revision=56406
- Desktop: Added code for server side switch over and some notes. Did not include desktopbg.c.
- Focus: Fixes wine Win tests for SetParent, Active, Focus and Foreground. Leaving the wine todos. Msg tests improved. Pop up menus are still broken see bug 6751. The tests reflect these pop up menu issues.
- WinPos: (WIP) Fixed MapWindowPoints. Arrange icons is still broken. SetWindowPos sets the foreground when it should.
- Implemented AllowSetForegroundWindow and LockSetForegroundWindow.
- Commit to reset base lines for RegTest ComCtl32 and related bugs.
- Coding credits and complements go out to the Wine team.
svn path=/trunk/; revision=56396
- Fix calculation of the brush origin. The coordinates passed to EngBitBlt "defines which pixel in the mask corresponds to the upper left corner of the destination rectangle." Our code was using it as corresponding to the surface origin.
The fix is currently only used when the new dib code is compiled. The old dib code is broken and expects the broken behaviour
svn path=/trunk/; revision=56392
- Troubleshooting bug 5630, moved DestroyProcessClasses from process to thread since process is called first. This way, it is assured the class destruction is on queue, but not before thread death.
- Added a global pointer to be used in a later commit.
- Warning: this does not fix bug 5630.
svn path=/trunk/; revision=56386
* Implement LoadSettings() and SaveSettings(). Patch by Edijs Kolesnikovics with some minor changes by me.
See issue #6925 for more details.
svn path=/trunk/; revision=56382
Fix a bug in MiAllocatePoolPages, that made the function succeed, when MAX_ULONG / -1 / 0xFFFFFFFF bytes were requested. The value overflowed into 0 and 0 pages were returned. When freeing this block, it could either free the next following large allocation or ASSERT when the end of the pool was reached without finding the end of the allocation.
Fixes FoxitReader 4.2/4.3
svn path=/trunk/; revision=56366
Mark _chkstk as extern to avoid duplication with msvcrtex library (using alloca will cause _alloca_probe to be linked, which aliases _chkstk)
svn path=/trunk/; revision=56361
- Implement saving / restoring debug registers on traps
- Replace the loop in KeContextToTrapFrame with something less ridiculous
- fixes a number of ntdd exception winetests
svn path=/trunk/; revision=56357
Expand the environment variables when running startup programs
Patch by Edijs Kolesnikovics, slightly modified by me
See issue #6953 for more details.
svn path=/trunk/; revision=56353
Add quotes to autocompleted names, when certain characters are present, not only spaces. Not sure if that's really all, but should be the most common.
svn path=/trunk/; revision=56350
Fix several bugs in the copy command:
- "x:" is not a valid source path (see bug #5709)
- when appending files using "+", cmd must distinguish between + as operator and + as part of the source file, when being put in "". For that reason the tokenizer was modified to make + a seperate token, which is replaced with | when creating the string of multiple files (since + is a valid file character)
- ? was not correctly handled when passed in the output file name
- factor out the code to create the file name from wildcards and make it readable (no, p, q, r are NOT reasonable variable names!)
- Move code out of the loop that doesn't belong there
- Make sure the file names are printed when multiple files are copied or appended
- write the count of written files (which is 1 when appending), not source files after the copy process is done
- Remove an excessive newline when writing strings with ConPuts()
- Improve readability a bit
svn path=/trunk/; revision=56346
Remove trailing backslash in shellink working dir
Patch by Edijs Kolesnikovics (terminedijs at yahoo dot com)
See issue #7000 for more details.
svn path=/trunk/; revision=56344
Fix path reference leaks in PATH_add_outline
Should fix failed assertion when closing google earth.
thanks to r3ddr4g0n for providing the neccessary debug info.
svn path=/trunk/; revision=56342
- Make the handle to the shutdown event a global handle and set it when the service manager is being shut down.
- Close the start event properly.
svn path=/trunk/; revision=56337
Implement RtlCreateBootStatusDataFile, RtlGetSetBootStatusData, RtlLockBootStatusData and RtlUnlockBootStatusData.
[SYSSETUP]
Call RtlCreateBootStatusDataFile as part of the install process.
svn path=/trunk/; revision=56332
* CMake automatically handles def files when they're added to the source list, so we don't need to mark it as an external object here.
svn path=/trunk/; revision=56321
- Don't include NDK from winnt.h, instead duplicate NtCurrentTeb inline definition between psfuncs.h and winnt.h as it used to be
- remove duplicated _GENERAL_LOOKASIDE_POOL
- Add missing NTKERNELAPI to ObDereferenceObjectDeferDelete
svn path=/trunk/; revision=56318
* Put ntoskrnl and ntdll in the same folder, because they both start with nt.
* Place hal next to acpi, they both deal with hardware stuff.
* Move mountmgr next to usb drivers.
* Put the rtl next to the crt, and share as much as possible between the two.
* I'm just kidding, you know, April's fool and all ;)
Here's the real deal:
[WIN32SS]
* And finally, integrate win32k itself into the new file layout. Now the subsystem modules are located close to each other (like gdi32 and ntgdi, or user32 and ntuser).
* Thanks go to the subsystem team for coming up with this nice layout, hopefully this will scale to the whole codebase layout soon.
* Special thanks to Smiley for coming up with this cool idea :)
svn path=/trunk/; revision=56311
* Initialize the subsystem restructure. Start by bringing-in gdi32 and user32. Brought to you by the awesome Subsystem Team.
svn path=/trunk/; revision=56305
- If neccessary unlock the working set in MiMakeSystemAddressValid before calling MmAccessFault, fixes a failed ASSERT, when coming from NtFreeVirtualMemory
- Remove duplicated ASSERT
svn path=/trunk/; revision=56300
- turkish translation of dxdiag by Arda Tanrýkulu (bug no. 6986);
- italian translation of browseui by Paolo Devoti (bug no. 6980);
svn path=/trunk/; revision=56298
* Make use of /client none and /server none to tell midl not to generate them.
* Add dummy proxy file generation in generate_idl_iids().
svn path=/trunk/; revision=56288
* Recognize the extra client/server files in add_rpc_files() that Mr. midl generates regardless of the passed flags, and mark them as dummy files so that they don't break the msvc parallel build.
svn path=/trunk/; revision=56286
- Don't rely on the current lookahead size for loopback packets. Instead, indicate the entire packet at once.
- The current lookahead size doesn't matter when indicating full packets. The packets are indicated in their entirety regardless of the lookahead size.
- The Parallels Tools NIC driver now works on ROS. Dedicated to Alex Ionescu.
svn path=/trunk/; revision=56285
- The legacy ROS Mm uses structures known as page ops to track operations on a page such as page out, page in, and access fault. The idea is that each operation is forced to wait until the page has completed all other pending operations to start its work. The problem was that the page op's completion event was a NotificationEvent instead of a SynchronizationEvent. This caused all operations to proceed at the same time if they were waiting on a single page op to begin. Now that is fixed and page ops proceed one after another as intended. This bug has been around since r9077.
- When waiting for sections to be unmapped or destroyed, the Mm waits for pending page operations to complete. The problem is that MmUnmapViewOfSection had a critical bug in which it forgot to dereference the page op it just retrieved. This caused zombie page ops to be stuck to that particular address if there were any pending operations at the time of the MmUnmapViewOfSection call. As a result, section destruction to bug check due to the hung page op after waiting 10 seconds for the operation to complete. This bug has been around since r18849.
- Due to the combined effects of the above bugs, sometimes unmapping or freeing a section would hang the system or bug check the system (remember the "Failed to wait for page op" messages?). This was evident in smiley_'s theme work when many applications calling FreeLibrary at the same time would trigger this bug. There are likely many more cases that triggered this series of events which either could never be replicated or simply could not be explained.
svn path=/trunk/; revision=56284
Attempt to fix a concurrency issue with midl. Regarding to the MSVC builder output, midl failed when compiling wine/itss.h due to a concurrent access to itss_i.c. Although the command to generate itss.h doesn't seem to geenrate itss_i.c (at least not on my wdk buld), I added a switch to rename that file.
svn path=/trunk/; revision=56280
"while (TRUE);" is probably the worst way of handling critical errors / unhandled failure pathes! Replace that with ASSERT(FALSE);
svn path=/trunk/; revision=56272
- Only initialize the nonpged pool after the pfn database was initialized
- "Fix" IRQL for the initialization of non paged pool
svn path=/trunk/; revision=56271
- Implement HwIdle() function to put cpu in idle mode, when waiting for keyboard input.
- Patch by Carlo Bramini (carlo dot bramix at libero.it)
See issue #6453 for more details.
svn path=/trunk/; revision=56270
Write a lot of general prose about the operation of the Mm parts of NewCC.
Dedicate to timo and chongo.
No code changes.
svn path=/trunk/; revision=56267
- romanian rapps translation update (including INF update) by Stefan Fulea (bug no. 6878);
- italian translation update by Paolo Devoti (bug no. 6980);
svn path=/trunk/; revision=56266
- ObReferenceObject() does NOT return an NTSTATUS. It returns the new reference count. Remove pointless NT_SUCCESS check
- Enable interrupts in amd54 page fault handler before calling MmAccessFault
- Add missing MmDecommittedPte for amd64
svn path=/trunk/; revision=56265
Mmake regedit to save/load settings for: window placement, listview columns size, last opened key, status bar, splitter.
Patch by Grégori Macário Harbs <mysoft64bits at gmail dot com> and Edijs Kolesnikovics <terminedijs at yahoo dot com>
See issue #6920 for more details.
svn path=/trunk/; revision=56256
Handle Ctrl-A in address bar.
Patch by Edijs Kolesnikovics (terminedijs at yahoo dot com)
See issue #5936 for more details.
svn path=/trunk/; revision=56255
Remove pointless ASSERTs, instead set the balance of the new node to balanced, before inserting it.
Dedicated to Alex.
svn path=/trunk/; revision=56253
- Refactor some multiply used overly complex expressions into human readable variable names
- Output some useful debuginfo when an import could bot be resolved
svn path=/trunk/; revision=56247
- Fix implementations of AcpiOsWaitSemaphore and AcpiOsAcquireMutex to obey the caller's demands to not block if requested
svn path=/trunk/; revision=56245
- Fix a stupid string comparison mistake that lead to false positives in the duplicate device detection (1 and 10 detected as conflict)
svn path=/trunk/; revision=56243
- Add *PNP0A08
[ACPI]
- Handle the PNP0A08 in a couple of missed cases
- Finally, PCI should work on systems that use the new PNP0A08 ID for identifying a PCI Express root bus
svn path=/trunk/; revision=56242
- Set DN_DISABLEABLE if the device can be disabled
[NEWDEV]
- Remove a hack needed because DN_DISABLEABLE was never set
svn path=/trunk/; revision=56241
- Add PNP0A08 as an alias for the PCI bus (more commonly known as PNP0A03)
- Should fix numerous PCI regressions on various machines since switching to ACPI (and the further regression of breaking boot since VBE is now PnP)
svn path=/trunk/; revision=56238
- Rewrite coordinate transformation code to use XFORMOBJ apis instead of using floating point in kernel mode. As a small present fix 14 gdi32_winetests (5 mapping, 9 bitmap)
I disabled some RtoL code, since it didn't work anyway and caused more bugs than it fixed.
svn path=/trunk/; revision=56236
- It appears ros wasn't ready for properly handling page faults with interrupts disabled, disable bugcheck and warn and dump stack instead for now.
- Make KeRosDumpStackFrameArray and KdbSymPrintAddress safe to use at elavated IRQL -- don't use Unicode formats for DbgPrint at possible elavated IRQL/interrupts disabled as this may generate a page fault. Safely convert to ANSI instead before printing out the address.
svn path=/trunk/; revision=56235
- Fix a crazy bug in KiTrap0EHandler, it didn't enable interrupts unless it trapped when interrupts were already disabled (during a page fault during page fault handling, for example). This seems to have worked because the old non-newcc version of MiReadPage appears to rely on a page fault to bring the paged out page in. Thanks Timo.
svn path=/trunk/; revision=56234
[NTOS]: Expand system space view buckets when they are running low.
[NTOS]: Support SEC_BASED section creates and mappings, implement based VAD parsing.
[NTOS]: Support section mappings at a fixed base address and check for conflicts.
[NTOS]: Define 8 prioritized standyby lists and initialize them. Also define the modified page list.
[NTOS]: Support mapping with SEC_COMMIT less than the entire size of the section.
[NTOS]: Detect and assert if ARM3 sections are attempted to be unmapped, since this isn't supported yet.
[NTOS]: Clean up some DPRINTs and clarify ARM3 ASSERTs vs. Windows ASSERTs.
svn path=/trunk/; revision=56232
- Implement proper use of alternate resource requirement lists
- Implement the case to set DN_NO_SHOW_IN_DM
- Implement reporting more problem codes: CM_PROB_NEED_RESTART, CM_PROB_FAILED_DRIVER_ENTRY, CM_PROB_FAILED_DRIVER_ENTRY, CM_PROB_DRIVER_FAILED_LOAD, CM_PROB_WILL_BE_REMOVED, CM_PROB_HELD_FOR_EJECT, CM_PROB_TRANSLATION_FAILED, and CM_PROB_NORMAL_CONFLICT
svn path=/trunk/; revision=56231
- Fix running RtlGetFullPathName_UstrEx test from drive root
- Add a test for RtlGetFullPathName_U based on the same test cases
svn path=/trunk/; revision=56230
- Fix right-to-left support the new EngBitBlt code
- Remove an ASSERT that we don't support
- Fix clipping in IntEngBitBlt by bailing out when the target rect got empty
- Delete obsolete file
You can use the code now, but there is a bug in the clipping code somewhere, sometimes causing the taskbar to be overwritten
svn path=/trunk/; revision=56224
- Add support for pattern brushes and bottom-to-top blt in new EngBitblt
- implement a new much simpler EngCopyBits
- improve clip enumeration code
The code now generally works very well now, only has a few bugs left
svn path=/trunk/; revision=56219
- Implement support for bottom up bitblt
- Add solid pattern support in generic BitBlt function
- Make sure right-to-left is only used in the appropriate functions
- Fix 1bpp _NextPixel() macro
svn path=/trunk/; revision=56218
- Do not overflow the uncommitted range descriptor segment. Really fixes notepad.
- Free the correct region in failure case
- Some comment fixes
See issue #6966 for more details.
svn path=/trunk/; revision=56217
- Add a trivial XLATEOBJ as a global variable.
- Make sure the XLATEOBJ is not NULL, when calling the diblib code
- Remove the IntEngMaskBlt stub, its a very special function for font rendering, keep the old one
- Add more driver debug assertions
- Remove obsolete definition
svn path=/trunk/; revision=56216
- Implement spacial even more optimized versions of SRCCOPY on equal surfaces for 8,16,24 and 32bpp.
- Add the right-to-left versions to the function tables
- Add _CALCSHIFT macro(s) to calculate the shift in the dib functions, instead of precalculating it in EngBitBlt. This costs us a few bytes per function (yes, with so many functions, every byte counts!) but since it's only for 1 and 4 bpp and improves the code by keeping DIB specific code out of the Eng function, it's reasonable to do so.
- Add optional optimization pragmas
svn path=/trunk/; revision=56212
- Apply fixes from Wine 1.4. Fixes invalid free in notepad and other edit control users
See issue #6966 for more details.
svn path=/trunk/; revision=56210
- Once again require a reboot for driver replacement since this seems to cause new issues with devices being enumerated multiple times on vmware
svn path=/trunk/; revision=56206
[NTOSKRNL]
- Bugcheck if a driver reports a PDO with a non-unique instance ID
[ACPI]
- Check that we're not reporting the same device multiple times
svn path=/trunk/; revision=56204
- Create PnpRootRegisterDevice which is used to register a given PDO with the root device so it won't report it twice
- Listing devices by connection works with PCI HAL now but it seems that ACPI is detecting multiple ACPI\PNP0C0F\0 instances here in vbox which is causing the same infinite loop
svn path=/trunk/; revision=56200
- Device manager now reports problem codes, shows unknown devices, and displays non-PnP drivers like Windows
- Inspired by Eric's recent work
[DEVMGMT]
- Finally (!!!) add support for showing devices with no drivers installed yet
- Expand the class tree if one of the devices has a problem
[INF]
- Add legcydrv.inf for installing the LegacyDriver class
[NTOSKRNL]
- Fix null termination issues and enable the legacy driver registry code
- Add a service description in legacy device entries
- Return valid flags for a device status query
- Store problem codes for several common device node failure states
svn path=/trunk/; revision=56199
Implement the 'View/Devices by Connection' feature. It runs very well on windows XP but fails on ReactOS. Needs to be investigated.
svn path=/trunk/; revision=56197
- As a side effect to 45140 one of the "non-used debug features", breaking in with WinDbg, would no longer work. Add this back, as well as the DPC timeout debug check and support for skipping ticks.
svn path=/trunk/; revision=56194
- Forgot this dereference
- Unloading works now, but there are many bugs in the device stacks that prevent it from occurring
- Usbhub will unload properly when a hub is removed for a driver update, but usbstor won't unload when all usb storage devices have been removed
- Work needs to be done to fix issues like this
svn path=/trunk/; revision=56185
- Reenable some dereferences of the driver object
- Fix return values in NtLoadDriver
- Split unloading into code for legacy drivers and non-legacy drivers
svn path=/trunk/; revision=56184
- Remove an extra deference to the device object that happened inside the IopDeleteDevice function (called by Ob when the reference count is 0)
svn path=/trunk/; revision=56183
Add pSetupGuidFromString(), pSetupIsGuidNull() and pSetupStringFromGuid() to the PSDK header setupapi.h and remove a local declaration.
svn path=/trunk/; revision=56182
Add InstallRoot key.
It is reg_sz and contains the full path to the installation directory.
Checked on Windows & Wine.
svn path=/trunk/; revision=56178
- Fix entrypoint and add startup code to fix registration of mlang classes. Fixes crashes and failures in inetcomm:mimeintl winetest and adds back many tests in mlang winetest
svn path=/trunk/; revision=56156
FsRtlIsDbcsInExpression/FsRtlIsNameInExpression: in case there are contigous stars in the expression, skip all the stars up to the last one and only consider the last one.
This fixes the failing tests.
svn path=/trunk/; revision=56143
- Add PCI-PCI bridges to the critical device database to fix discovery of critical devices that are not attached directly to the root bus
- See bug #6958 for details
svn path=/trunk/; revision=56140
- When installing dll copy manifest to winsxs\manifests directory as comctl32 before sync did. Fixes Firefox 10 regression.
See issue #6910 for more details.
svn path=/trunk/; revision=56132
- Fix broken VIDEO_PORT_GET_DEVICE_EXTENSION macro
- Add support for child devices (monitors only for now)
- Monitors now show up in device manager ("Plug and Play Monitor" shows up if EDID data is reported, otherwise "Default Monitor" is displayed)
svn path=/trunk/; revision=56128
- It appears that the driver unloading code is more profoundly broken that I had previously thought. I ran into issues testing with videoprt which would call IoCreateDevice and IoDeleteDevice during enumeration of child devices. The issue arose when the IoDeleteDevice call unloaded the video driver which was currently initializing. This caused video initialization to fail and the system to shutdown.
- Disable this code completely for now until the proper algorithm for determining when a driver should unload is found
svn path=/trunk/; revision=56125
32-Bit stub installer for 16-Bit Microsoft installers
TODO: -load this installer instead of the original 16bit Microsoft installer
-improve error handling and string allocation
svn path=/trunk/; revision=56110
* Make use of the cmake rc handling. As a result we don't need set_rc_compiler() anymore, among other benefits.
* There's one minor issue mentioned as a FIXME, we'll get that sorted out as soon as the bug report we filed upstream gets resolved.
* Set set_rc_compiler() to be a dummy, to be removed later.
[IMM32]
* Properly set the compiler flag.
svn path=/trunk/; revision=56105
Handle cases where multiple ranges overlap a new lock, including either exclusive or shared
ranges. We expand a shared range when an overlap occurs and add another shared range to
the shared ranges list. Also make sure to complete an IRP along every non-pending case.
Reduces failures in kernel32_winetest except an expected successful unlock of ~0,1, which
appears to contravene the documentation.
svn path=/trunk/; revision=56098
- Dereference the correct device object (the one who asked for shutdown notification) instead of whatever happens to be on the top of the device stack
- Fix flag issues
svn path=/trunk/; revision=56087
* unify spec2def and add_importlib_target
* use module name instead of spec file name for deciding importlib target name
svn path=/trunk/; revision=56081
- Improve drive properties dialog. Patch by Edijs Kolesnikovičs <terminedijs at yahoo dot com>
See issue #6833 for more details.
svn path=/trunk/; revision=56075
- Improve file properties dialog. Patch by Edijs Kolesnikovičs <terminedijs at yahoo dot com>
See issue #6829 for more details.
svn path=/trunk/; revision=56074
- When using output rect clipping results in input rect, scale them to input rect size. Fixes gdiplus:graphics winetest crash.
svn path=/trunk/; revision=56072
- Fix the unload path
- This does expose some bugs: 2 cont-able assertions in ARM3 after unplugging a USB storage device (during usbstor unload), HID unload doesn't seem to work correctly (somebody still has references to the devices), other issues may pop up too
svn path=/trunk/; revision=56069
- Fix several driver object reference leaks
- Fix a kmtest regarding previously loaded drivers
- Part 1 of getting PnP driver unload to work properly
svn path=/trunk/; revision=56063
- Fix the way we undefined IoCallDriver and IoCompleteRequest so existing and future code in irp.c will use the correct function when using the Io* versions.
Part 1 of 2.
svn path=/trunk/; revision=56060
- Assert that the PFN lock is held during MmReferencePage
- Assert that the PFN lock is not held during MmFreeMemoryArea
svn path=/trunk/; revision=56052
- Fix incorrect placement of the PDE freeing code which caused a deadlock to occur if a low memory condition was indicated by ARM3
svn path=/trunk/; revision=56050
Part 1: Page Fault Path Fixes
[NTOS]: As an optimization, someone seems to have had changed the MiResolveDemandZeroFault prototype not to require a PTE, and to instead take a protection mask directly. While clever, this broke support for ARM3 sections, because the code was now assuming that the protection of the PTE for the input address should be used -- while in NT Sections we instead use what are called ProtoType PTEs. This was very annoying to debug, but since the cause has been fixed, I've reverted back to the old convention in which the PTE is passed-in, and this can be a different PTE than the PTE for the address, as it should be.
[NTOS]: Due to the reverting of the original path, another optimization, in which MiResolveDemandZeroFault was being called directly instead of going through MiDispatchFault and writing an invalid demand-zero PDE has also been removed. PDE faults are now going through the correct, expected path.
[NTOS]: MiResolveDemandZeroFault was always creating Kernel PTEs. It should create User PTEs when necessary.
[NTOS]: MiDeletePte was assuming any prototype PTE is a forked PTE. Forked PTEs only happen when the addresses in the PTE don't match, so check for that too.
Part 2: ARM3 Section Object Fixes
[NTOS]: Fix issue when trying to make both ROS_SECTION_OBJECTs and NT's SECTION co-exist. We relied on the *caller* knowing what kind of section this is, and that can't be a good idea. Now, when the caller requests an ARM3 section vs a ROS section, we use a marker to detect what kind of section this is for later APIs.
[NTOS]: For section VADs, we were storing the ReactOS MEMORY_AREA in the ControlArea... however, the mappings of one individual section object share a single control area, even though they have multiple MEMORY_AREAs (one for each mapping). As such, we overwrote the MEMORY_AREA continously, and at free-time, double or triple-freed the same memory area.
[NTOS]: Moved the MEMORY_AREA to the "Banked" field of the long VAD, instead of the ControlArea. Allocate MMVAD_LONGs for ARM3 sections for now, to support this. Also, after deleting the MEMORY_AREA while parsing VADs, we now use a special marker to detect double-frees, and we also use a special marker to make sure we have a Long VAD as expected.
svn path=/trunk/; revision=56035
- Give freeldr its own RtlAssert to avoid linking to the one in rtl. Makes it possible to use the DbgPrompt version (currently #if 0-ed) of RtlAssert in ntdll/ntos without breaking freeldr linking.
svn path=/trunk/; revision=56028
Fix my lock implementation to a degree.
- Have a single list associated with the CSQ.
- Remove some confusion from when the lock information was a
generic table.
- Make the compare function not be confused by attempts to
give strange offsets and lengths.
- Make the uninitialize function complete all lock requests
as specified.
- Remove my unneeded attempt to shoehorn in an eresource.
svn path=/trunk/; revision=56026
Fix paging out.
- We know we're in a wait entry in the section. What we really want to know
is if it's safe to evict this address space when withdrawing the mappings.
- We don't want to corrupt our wait entry in the section page table to keep
accumulating dirty state into the page out process until it's time to finally
evict.
svn path=/trunk/; revision=56024
- Add more items to taskbar popup menu. Patch by Edijs Kolesnikovičs <terminedijs at yahoo dot com>
See issue #6242 for more details.
svn path=/trunk/; revision=56017
- Fix a major bug in the failure path of IoCreateDevice that would cause corruption of kernel memory if IopCreateVpb failed
svn path=/trunk/; revision=56014
- Fix folder properties dialog. Patch by Edijs Kolesnikovičs <terminedijs at yahoo dot com>
See issue #6832 for more details.
svn path=/trunk/; revision=56013
- Improve shortcut properties dialog. Patch by Edijs Kolesnikovičs <terminedijs at yahoo dot com>
See issue #6848 for more details.
svn path=/trunk/; revision=56011
- Using a dead stack lookaside list for up to 5 dead kernel thread stacks. 1500% improvement when rapidly destroying/creating threads, such as during second stage setup and many winetests.
- Using a free pool page lookaside list for up to 12 free non-paged or paged pool pages. 800% improvement when allocating big pages from the pool, as well as during pool expansion.
- Using a bucketized per-processor and local list (in the KPRCB) for block sizes between 1 and 32 bytes. 1000% improvement when rapidly allocating/freeing small pool allocations, and 8x reduction in pool fragmentation.
svn path=/trunk/; revision=56000
- Fix a possible integer overflow in NtGdiCreateBitmap
- Don't cast length of LARGE_STRING (ULONG) to USHORT, instead truncate to MAXUSHORT. This is still hacky, but better than before.
- Fix MSVC warnings
svn path=/trunk/; revision=55993
- Fix bug while scanning endpoint descriptors
- Do not assume interface info has the correct number of pipes set
- Fix alternate interface handling
- Tested in VBox + USB Audio Device
svn path=/trunk/; revision=55992
[NTOS]: The fix above was due to fixing "EndingAddress" which was being initialized to zero too late (after writing to it!). This caused allocations with a fixed base address that were already on top of another allocation not to be seen as a conflict, then we tried inserting a VAD and received an ASSERT saying we've already found a VAD there. After fixing the sizing code, the bug above creeped up.
Whoever wrote the NtFreeVirtualMemory test is a godsend. It has been nailing bug after bug in the VAD implementation. Thank you.
svn path=/trunk/; revision=55990
- Back by popular demand, the storage driver to end all storage drivers, uniata!
- I had forgotten how many issues atapi had itself and the hardware compatibility improvement (if there was one) was offset by lack of features
svn path=/trunk/; revision=55978
Hoping this will work on other configurations other than just my test virtual machine. If so, it is a big day in Mm history. Only sections and cache still use the old Mm functionality. I -will- revert this if all hell breaks loose.
svn path=/trunk/; revision=55977
- Implement support for getting string descriptors (may need more work)
- USB mass storage devices which are part of a composite device now initialize
svn path=/trunk/; revision=55976
When IoReportDetectedDevice() creates a PDO, the PDO is in no way related to the DriverObject given to the function.
Driver calling IoReportDetectedDevice() should then ensure that it attachs a FDO on top of this PDO.
svn path=/trunk/; revision=55972
- Don't assert that the clear stall will complete successfully since it may be sent for a removal
- Fixes removal surprise removal of HID devices
svn path=/trunk/; revision=55971
- Rewrite broken USBD_CreateConfigurationRequestEx
- It did not calculate the required urb size correctly
- It did not store the correct number of provided interfaces
- Length of each interface information was wrong
svn path=/trunk/; revision=55965
- Experimental change to try to increase real hardware compatibility
- Load both atapi and uniata and allow atapi to have priority and uniata will pick up devices that atapi doesn't like
- Fix a bug in uniata enumeration
svn path=/trunk/; revision=55958
-it looks like IF(${CMAKE_SYSTEM_NAME} MATCHES "Windows") return TRUE on MacOS/Linux when cross compiling
-file(TO_NATIVE_PATH) is broken as well
...
next try
svn path=/trunk/; revision=55954
- Implement enabling extended debug output in freeldr without the need to recompile it. Add an option in the f8 menu that allows the user to specify which debug channels and levels will be enabled.
svn path=/trunk/; revision=55946
- Split retrieving the full configuration descriptor into 2 steps
- Fixes errors while initializing usb devices on UHCI controller
- Tested in VmWare / VBox and real hardware
svn path=/trunk/; revision=55942
1) Assert on empty kernel PTE instead of handling it as a bugcheck. Windows ASSERTs too. Also clarify some ASSERTs which Windows also does versus ASSERTs we are only doing due to lack of support for said feature.
2) User page fault path can now distinguish between a user-mode PTE fault, and a kernel-mode fault on a user PDE, both by creating a correct kernel PDE when needed instead of always creating user PTEs, as well as by only touching the UsedPageTableEntry reference counting mechanism when a user-address is in play.
3) Related to #2, also recognize when the faulting PTE is actually a PDE in the self-mapping region -- another scenario when the "user fault" is actually a kernel fault for a user PDE.
4) Add one more path where a Paged Pool PDE fixup can save the day instead of always faulting.
5) Finally, related to #2 and #3, handle the MI_IS_PAGE_TABLE_OR_HYPER_ADDRESS scenario for a User PDE by treating it as a user fault. The code looks deceptively similar but there are slight differences which require the separate codepaths with some duplicated code. The magic is in the ordering.
In trunk, these changes should not cause any regressions (let's hope so). On the internal VAD-based Virtual Memory branch, they now allow booting to 3rd stage and a fully usable ReactOS environment. MEMORY_AREA_VIRTUAL_MEMORY is gone on that branch. It's coming.
[NTOS]: Use PAGE_READWRITE as hardcoded protection instead of PAGE_EXECUTE_READWRITE -- the difference is meaningless on ReactOS Mm but actually causes issues on ARM3 with VADs.
svn path=/trunk/; revision=55938
- Add a missing device node flag to legacy reported devices to avoid an assertion failure during installation
- The fact that this was exposed by a change in device node linking order terrifies me
svn path=/trunk/; revision=55926
- assert PDE ref count consistency everywhere
- fix the fix(tm) : after reference decrement, the maximum must not be reached
svn path=/trunk/; revision=55925
- Clean up code, fix memory leaks, check returns codes, add asserts
- Use root device handle which is is prerequisite for usb hub support
[USBLIB]
- Fix root hub handle checks
- Add more code for hub support
svn path=/trunk/; revision=55923
- Fix linking of device children to respect the enumeration order instead of linking in reverse enumeration order
- PCI cards (and other devices) now enumerate in the correct order
svn path=/trunk/; revision=55912
- Don't modify PCI configuration in AddDevice function
[USBEHCI]
- Don't halt the controller while the BIOS still has control of it
svn path=/trunk/; revision=55910
- Change MmDereferencePage to MmReleasePageMemoryConsumer to do proper page usage tracking
- Remove an unnecessary reference and dereference
svn path=/trunk/; revision=55897
- Fix formatting FAT or NTFS volumes into EXT2 by defining ZAP_BOOTBLOCK and fixing a bug in zap_sector()
[USETUP]
- Enable ext2 as an option in setup
- Formatting and building file copy list works but setup stalls shortly after starting to copy files (NewCC on)
svn path=/trunk/; revision=55896
- Fix several locking issues to better fit with RosMm
- Page out is critical as we use it now (shutdown) so force address space locking
- Fix an address space lock leak for VM regions
- Quiet some debugging and enable/change some other
- Still some lingering issues in page out exist which seem to be related to zefklop's PDE ref counting stuff and a weird one where we fault on the PFN database
- Besides the aforementioned issues, NewCC is surprisingly solid. I was able to boot to 3rd stage without a problem and run some apps.
svn path=/trunk/; revision=55895
1) MiCheckVirtualAddress should be called *after* determining if the PTE is a Demand Zero PTE. This is because when memory is allocated with MEM_RESERVE, and then MEM_COMMIT is called later, the VAD does not have the MemCommit flag set to TRUE. As such, MiCheckVirtualAddress returns MM_NOACCESS for the VAD (even though one is found) and the demand zero fault results in an access violation. Double-checked with Windows and this is the right behavior.
2) MiCheckVirtualAddress now supports non-commited reserve VADs (ie: trying to access MEM_RESERVE memory). It used to ASSERT, now it returns MM_NOACCESS so an access violation is raised. Before change #1, this would also happen if MEM_COMMIT was later performed on the ranges, but this is now fixed.
3) When calling MiResolveDemandZeroFault, we should not make the PDE a demand zero PDE. This is senseless. The whole point is that the PDE does exist, and MiInitializePfn needs it to keep track of the page table allocation. Removed the nonsensical line of code which performed cleard the PDE during a demand-zero fault.
I am able to boot to 3rd stage with these changes, so I have seen no regressions. Additionally, with these changes, the as-of-yet-uncommitted VAD-based Virtual Memory code completes 1st stage setup successfully, instead of instantly crashing on boot.
svn path=/trunk/; revision=55894
* The default build will continue to include all the localizations, but you can set it otherwise through a cmake variable (I18N_LANG) at configure time.
* Brought to you by Team Localization (Kamil Hornicek and Amine Khaldi) ;)
svn path=/trunk/; revision=55891
Disable tests that would hit asserts due to current work on VM. Requested functionality is missing and will be for some time.
svn path=/trunk/; revision=55888
[NTOS]: Do a more stringent check to refuse COPY_ON_WRITE flag sent through NtAllocateVirtualMemory.
[NTOS]: For VM-based Memory Areas, acquire and then release the process working set lock while inserting the VAD, to simulate what the VAD-based Virtual Memory behavior will look like.
[NTOS]: Disable paging for VM-based Memory Areas since this will not be supported with VADs.
[KERNEL32]: CopyLoop was requesting 2 zero bits when calling NtAllocateVirtualMemory. Not sure if this was really the intent or not, but both the new as well as the old NtAllocateVirtualMemory do not support this (the new one will ASSERT). Since this functionality never worked, request 0 bits instead to avoid hitting the ASSERT.
Any problems with what the VAD system will introduce should be revealed by now. From this build until the one which will have the VAD-based system in place, no further VM-related issues should crop up.
svn path=/trunk/; revision=55885
[NTOS]: Implement counters for paged and nonpaged pool. First-stage setup now shows Kernel Pool values again, as does Task Manager. Fixes the regression introduced when pool/non-paged pool was no longer managed through "memory consumer" API.
[NTOS]: Add more debugging paths and flags that were sent over from the "Aleksey Pool Patch". Most of them are not implemented.
[NTOS]: Fix a missing case when a pool header check was not being done.
[NTOS]: Check IRQL levels during pool allocation and free.
With tags and the counters, pool leaks should be massively more debuggable now.
svn path=/trunk/; revision=55879
- Fix declaration of HalPrivateDispatchTable for FreeLDR.
[FREELDR]
- Rename BiosMemoryMap to XboxMemoryMap. Another BiosMemoryMap is in meminit.c
[NTOSKRNL][WIN32K][FTP]
- Fix some variables.
svn path=/trunk/; revision=55878
* Introduce a way to compile GCC builds with PDB style debug info (isolated into .dbg files). This will reduce the memory footprint considerably, among other benefits.
svn path=/trunk/; revision=55873
- null terminate DesktopObject->pDeskInfo->szDesktopName and WindowStationObject->Name
- null terminate the buffer when NtUserGetObjectInformation is called with UOI_NAME parameter
- should fix a random failure in user32:desktop tests
svn path=/trunk/; revision=55857
- Create a macro for properly checking for low speed devices
- Check again for a low speed device after the EHCI reset completes
- Don't ignore status changes on ports that we have released (revert part of r55851)
svn path=/trunk/; revision=55853
- Fix reset event handling in cases where reset does not end up in a new device created
- My MacBook Pro completes USB initialization free of hangs now (usbstor disabled)
svn path=/trunk/; revision=55852
- Handle errors returned from the HCD correctly
[USBEHCI]
- Discard any changes on ports that were given to the companion controller
svn path=/trunk/; revision=55851
- Fix reset port sequence #2
- Clear unwanted bits
- Fix wait time as in spec
- Wait max 500ms for the reset bit to clear, otherwise restart the reset sequence
- Check if work item is currently queued
svn path=/trunk/; revision=55849
- Fix the Live CD regression (RBuild only - The irony is not lost on me) caused by r55555
- There is another regression that still prevents booting though
svn path=/trunk/; revision=55846
- Properly create the WindowStations directory in NtUserInitialize. Now we take into account the fact that if we are not in session 0, it should be created in \Sessions direcory
- Clean up IntParseDesktopPath and make it use the correct WindowStations
- Add some comments and cleanup
- Remove the unused IntGetFullWindowStationName, IntGetWindowStationObject and IntGetWinStaObj
svn path=/trunk/; revision=55841
- Remove nasty hack to make Windows PCI driver load. This may not be required after the fixes that were committed after this hack. If it is still necessary then please fix whatever issue makes the PCI driver crash instead of relying on a return from bugcheck call to work...
[DDK]
- Mark KeBugCheckEx as noreturn again. Should fix Prefast issues where it thought execution could return from the call and would warn about false issues.
svn path=/trunk/; revision=55840
- Specify C4700 (uninitialized variable usage) to cause an error on compile as this is very likely to result in broken code.
[NTDLL]
- Disable ASSERT in LdrpSearchPath that could be reading an uninitialized variable. The routine is not fully implemented so this should likely be re-enabled once it is.
[BROWSEUI]
- Add missing call to BuildRebarBandInfo, or we would end up with an unitialized variable sent to SendMessage.
[PCIX]
- Fix broken ASSERT
- Fix usage of wrong (and uninitialized) variable in the Unicode case for PciGetDescriptionMessage.
[DDK]
- Prepend "_" to local variable of NdisChainBufferAtBack. Fixes code in ndisuio that called this using the same name as the macro's variable (which resulted in NdisChainBufferAtBack assigning the variable to itself).
[WIN32K]
- Fix WinPosShowIconTitle that checked if a variable was NULL before assigning it to anything.
svn path=/trunk/; revision=55834
A reintegration checkpoint for the NewCC branch, brought to you by Team NewCC.
Differences with current ReactOS trunk:
* A new memory area type, MEMORY_AREA_CACHE, is added, which represents a mapped region of a file. In NEWCC mode, user sections are MEMORY_AREA_CACHE type as well, and obey new semantics. In non-NEWCC mode, they aren't used.
* A way of claiming a page entry for a specific thread's work is added. Placing the special SWAPENTRY value MM_WAIT_ENTRY in a page table, or in a section page table should indicate that memory management code is intended to wait for another thread to make some status change before checking the state of the page entry again. In code that uses this convention, a return value of STATUS_SUCCESS + 1 is used to indicate that the caller should use the MiWaitForPageEvent macro to wait until somebody has change the state of a wait entry before checking again. This is a lighter weight mechanism than PAGEOPs.
* A way of asking the caller to perform some blocking operation without locks held is provided. This replaces some spaghettified code in which locks are repeatedly taken and broken by code that performs various blocking operations. Using this mechanism, it is possible to do a small amount of non-blocking work, fill in a request, then return STATUS_MORE_PROCESSING_REQUIRED to request that locks be dropped and the blocking operation be carried out. A MM_REQUIRED_RESOURCES structure is provided to consumers of this contract to use to accumulate state across many blocking operations. Several functions wrapping blocking operations are provided in ntoskrnl/cache/reqtools.c.
* Image section pages are no longer direct mapped. This is done to simplify consolidation of ownership of pages under the data section system. At a later time, it may be possible to make data pages directly available to image sections for the same file. This is likely the only direct performance impact this code makes on non-NEWCC mode.
RMAPs:
* A new type of RMAP entry is introduced, distinguished by RMAP_IS_SEGMENT(Address) of the rmap entry. This kind of entry contains a pointer to a section page table node in the Process pointer, which in turn links back to the MM_SECTION_SEGMENT it belongs to. Therefore, a page belonging only to a segment (that is, a segment page that isn't mapped) can exist and be evicted using the normal page eviction mechanism in balance.c. Each of the rmap function has been modified to deal with segment rmaps.
* The low 8 bits of the Address field in a segment rmap denote the entry number in the generic table node pointed to by Process that points to the page the rmap belongs to. By combining them, you can determine the file offset the page belongs to.
* In NEWCC mode, MmSharePageEntry/UnsharePageEntry are not used, and instead the page reference count is used to keep track of the number of mappings of a page, allowing the last reference expiring to allow the page to be recycled without much intervention. These are still used in non-NEWCC mode. One change has been made, the count fields have been narrowed by 1 bit to make room for a dirty bit in SSE entries, needed when a page is present but unmapped.
Section page tables:
* The section page tables are now implemented using RtlGenericTables. This enables a fairly compact representation of section page tables without having the existence of a section object imply 4k of fake PDEs. In addition, each node in the generic table has a wide file offset that is a multiple of 256 pages, or 1 megabyte total. Besides needing wide file offsets, the only other visible change caused by the switch to generic tables for section page tables is the need to lock the section segment before interacting with the section page table.
Eviction:
* Page eviction in cache sections is accomplished by MmpPageOutPhysicalAddress. In the case of a shared page, it tries to remove all mappings of the indicated page. If this process fails at any point, the page will simply be drawn back into the target address spaces. After succeeding at this, if TRUE has been accumulated into the page's dirty bit in the section page table, it is written back, and then permanently removed.
NewCC mode:
* NEWCC mode is introduced, which rewrites the file cache to a set of cache stripes actively mapped, along with unmapped section data.
* NewCC is more authentic in its interpretation of the external interface to the windows cache than the current cache manager, implementing each of the cache manager functions according to the documented interface with no preconceived ideas about how anything should be implemented internally. Cache stripes are implemented on top of section objects, using the same memory manager paths, and therefore economizing code and complexity. This replaces a rather complicated system in which pages can be owned by the cache manager and the memory manager simultaneously and they must cooperate in a fairly sophisticated way to manage them. Since they're quite interdependent in the current code, modifying either is very difficult. In NEWCC, they have a clear division of labor and thus can be worked on independently.
* Several third party filesystems that use the kernel Cc interface work properly using NEWCC, including matt wu's ext3 driver.
* In contrast with code that tries to make CcInitializeCacheMap and CcUninitializeCacheMap into a pair that supports reference counting, NEWCC lazily initializes the shared and private cache maps as needed and uses the presence of a PrivateCacheMap on at least one file pointing to the SharedCacheMap as an indication that the FILE_OBJECT reference in the SharedCacheMap should still be held. When the last PrivateCacheMap is discarded, that's the appropriate time to tear down caching for a specific file, as the SharedCacheMap data is allowed to be saved and reused. We honor this by making the SharedCacheMap into a depot for keeping track of the PrivateCacheMap objects associated with views of a file.
svn path=/trunk/; revision=55833
- Delay for the correct amount of time when performing a global reset and a port reset
- Loosen the loop timings a bit to allow for more time for real hardware to complete start/stop/reset
- Clear only the change bits that were set and handle both bits in the either case
- Perform a synchronous callback when the reset completes
svn path=/trunk/; revision=55832
- OHCI -> UHCI
- Implement timer routine which checks for new detected devices. In case a port change is detected, the sce callback routine is invoked
svn path=/trunk/; revision=55828
- Reenable and fix the correct reset code
[HIDCLASS]
- Disable removal of PDOs due to IRP cancellation issues
[USBEHCI][USBOHCI][USBUHCI]
- Disable removal of FDOs due to reinitialization issues
svn path=/trunk/; revision=55827
- Print interface details when dumping the function descriptor
- Fix bug USBCCG_AppendInterfaceNumber, which is not yet used
svn path=/trunk/; revision=55825
Some misc fixes to make kernel compile warning-free with /W3. Also fix a bug that completely broke the ProcessWx86Information case for NtQueryInformationProcess by missing to reference/dereference and acquiring the specified process' EPROCESS.
svn path=/trunk/; revision=55815
- Do not depend on argument evaluation order and thus uninitialized values. Sent and applied upstream. Dedicated to Stefan
svn path=/trunk/; revision=55814
[DDK]
- ndis.h Don't use local variables with the same name as function parameters.
[PSDK]
- access.idl Remove incorrect annotation
- usbioctl.h Not really a Prefast issue, but switch to use ANYSIZE_ARRAY like we do everywhere else.
- winddi.h EngQueryDeviceAttribute's 6th parameter is IN, not OUT
- winefs.h Fix FileEncryptionStatus to use annotations. Its definition in winbase.h already has them and this caused a conflict.
[WINE]
- debug.h Remove unnecessary check (in ReactOS) that made Prefast sad.
- test.h Disable a false-positive C28182.
svn path=/trunk/; revision=55813
- Detect the size of the configuration descriptor before obtaining the full configuration descriptor
- Fix integer overflow in BuildTransferDescriptorChain
svn path=/trunk/; revision=55810
- Queue dpc when the interrupt indicates completion of a transfer or an error interrupt
- Implement checking if a queue head is complete
- Free queue heads and associated endpoint descriptors
svn path=/trunk/; revision=55809
- Write the configure bit
- Clear connection change and port suspend bit when set after controller initialization
- Queue work item after the reset port is complete
- Disable port before reseting it
- Wait time for reset is 50ms, not 25ms
- Clear reset change bit when clearing reset feature
- Implement enabling port feature
svn path=/trunk/; revision=55806
- Fix use of uninitialized variables in failure case of RtlSetCurrentDirectory_U. CID 15339, 15340
- Handles should be NULL, not 0
svn path=/trunk/; revision=55805
- When deleting an old page file, take the newly freed space into account
- Keep at least 64 MB of disk free instead of 32 so that 2nd stage can succeed with a small disk
- Miscellaneous fixes and simplifications in the page file code
svn path=/trunk/; revision=55803
[KERNEL32]: Enable support for asynchronous I/O callbacks with SxS/Fusion activation context support. Right now only added to the file Change notification APIs, but ReadFileEx and WriteFileEx should use them too.
svn path=/trunk/; revision=55801
- Properly destroy the THREADINFO if we fail to create it. We don't leak the THREADINFO no in case of failure. It should also fix random assertion failure when running user32:desktop tests
svn path=/trunk/; revision=55798
- C standard for free says "if pointer is NULL, no action is taken".
- MSDN for HeapFree says "if pointer is NULL, behavior is undefined".
Thanks to Thomas for teaching wine guys how to read.
svn path=/trunk/; revision=55797
[KERNEL32]: Lock/UnlockFile APIs should handle STATUS_PENDING/ERROR_IO_PENDING situations instead of always assuming the Nt functions return synchronously.
[KERNEL32]: Other stylying changes + commenting/annotations.
svn path=/trunk/; revision=55795
- Implement function to retrieve queue head for the specified transfer type
- Pass device speed to IUSBRequest initialization routines
- Implementing inserting the new queue head into the required queue head
- Implement support for control transfers, not yet used
svn path=/trunk/; revision=55787
- allocate FAST_MUTEX objects from non paged pool.
This should ix a bunch of weird testbot failures.
Any suggestion on the TAG is welcome
svn path=/trunk/; revision=55784
- generate 'real' libraries for import libraries
This way only the link step relies on them.
Should avoid rebuilding half of the tree next time Alex changes ntdll.spec
A relinking orgy is more than enough
svn path=/trunk/; revision=55776
- Add prints to unloading functions because things seem very wonky here
- This exposes some previously unknown behavior of fastfat actually being unloaded in early boot (with uniata and buslogic failing unload due to missing DriverUnload)
svn path=/trunk/; revision=55773
- Call mini driver for pnp events
- Wait for all pending irps to complete
[HIDUSB]
- Fix bug when removing device object found by Cameron Gutman
[USBOHCI]
- Remove assert
- Delete configuration descriptors when device is deleted
svn path=/trunk/; revision=55772
- finally, release user shared data at process address space cleanup.
- release PDE pages that might not be freed at process end.
- Let the caller handle PDE release when deleting a PTE
- restore Richard's ASSERT : All user PDE pages are now freed!
svn path=/trunk/; revision=55770
* Alter the current base addresses, and introduce all the missing ones.
* We have a much better way to assign them, thanks to Timo's excellent suggestions.
* Big thanks go to Geir Hauge for making us an awesome script that conveniently and automagically generates baseaddress.cmake.
* Dedicated to Alex.
svn path=/trunk/; revision=55767
- Do not store the PDEVOBJ and a fake name in MONITOR struct. Use MONITOR.HDEV instead (which is an opaque pointer to a PDEVOBJ)
svn path=/trunk/; revision=55755
- "All this file is a big hack and should be removed one day…" - Today is that day! setup.c is no more!
[TXTSETUP.SIF]
- Move keyboard drivers to the Keyboard.Load section
[USETUP]
- Add a small keyboard class installation hack
- USB keyboards now load during 1st stage
svn path=/trunk/; revision=55750
- Create the DeviceClasses key necessary for device interfaces to be registered successfully in 1st stage
- This may not be the correct place to do this so anyone is free to move it
svn path=/trunk/; revision=55747
- When releasing a low or full speed device, set the EHCI_PRT_RELEASEOWNERSHIP bit with the current port status so that any change bits are acknowledged
- Fixes detecting new devices after a companion controller handoff
svn path=/trunk/; revision=55742
[NTOSKRNL]: If you're going to fake success when doing access checks, also grant the rights that would normally be refused, for consistency's sake. Should fix a bunch of the kernel32:synch tests (which have nothing to do with synch...).
svn path=/trunk/; revision=55737
[NTOSKRNL]: The "Write" bit in the x86 Error Code is not bit 1, it is bit 2. Fix all the GPF handlers to check for bit 2 instead when determining if an exception was due to read or write. Fixes Winetests.
I'll say it nicely this time for Timo: when you rewrite ASM, do it carefully :)
svn path=/trunk/; revision=55735
[NTDLL/RTL]: Fix RtlIpv4AddressToStringA and RtlIpv4AddressToStringW to work like in Windows (and crash with a NULL buffer).
svn path=/trunk/; revision=55729
- remove assertions from the process callout checking that ppiCurrent->rpdeskStartup and ppiCurrent->hdeskStartup are valid. This can happen in case we fail to open the initial desktop or window station
svn path=/trunk/; revision=55726
- Try to fix reset bugs in my code and remove hacks
- Don't clear extra bits when acknowledging a port connect status change
[USBOHCI]
- Code cleanup
- No functional change
svn path=/trunk/; revision=55725
- Fix a crash in KiSystemCall that was caused because PsConvertToGuiThread may have switched to a large kernel stack, but still returned with failure because win32k did not succeed. To fix it reload the trap frame after the call before checking for success.
svn path=/trunk/; revision=55724
[CSRSRV]: Add special handling for certain messages (similar to how "Reply" is used in CSRSRV2) so that we don't keep replying to the wrong thead. For example, during an ExitProcess, special handling is needed at the end of the loop.
[CSRSRV]: Remove the last few remaining legacy process management function and use the correct CsrLock/Unlock APIs instead. No longer leak references for every process like crazy. Fully manage proc/thread/session reference counts now.
At this point it looks like there might just be one issue left before this thing works right (minus potential regressions due to missed hacks).
svn path=/trunk/; revision=55718
[CSRSRV/kERNEL32]: Switch to using the Windows CSRSS message structure for a new process, minus the bInheritHandles field which we still support for ReactOS compatibility.
[CSRSRV]: Port CsrCreateProcess from CSRSRV2, rewrite CsrSrvCreateProcess to use this API, and add required ReactOS compatibility hacks. A lot of stuff is still not supported, but the minimum (ie: previous ReactOS functionality) has been maintained during the transition. Processes are now tracked correctly, but there are still refcount leaks (known, to be fixed later).
Will let this breathe for a bit now to catch regressions. Work can now continue back on the kernel32 side while that happens.
svn path=/trunk/; revision=55711
[NTOSKRNL]: Implement ProcessUserModeIOPL info level (and implement Ke386SetIopl) instead of spamming we can't do this.
[NTOSKRNL]: Implement ProcessExecuteOptions info level (and implement MmSetExecuteOptions) instead of spamming we can't do this.
[NDK]: Add NoExecute Flags based on ProcessHacker.
No longer spammed to death for every process all the time.
svn path=/trunk/; revision=55688
- Don't preserve command status values when notifying controller of a new bulk / control endpoint
- Abort pipe & reset data toggle in the sync reset routine
- Allocate interface descriptors and endpoint handles when creating the configuration descriptor
- Implement routine for allocating chained descriptors
- Implement data toggle for bulk&interrupt transfers
- Mass storage devices should now work in real hardware with OHCI controller
- Tested in real hardware with NEC Corporation USB [1033:0035] (rev 43)
svn path=/trunk/; revision=55687
KdInitSystem may be called after system set up if /CRASHDEBUG is enabled and a bugcheck or NMI occurs. With the INIT discard/delete now implemented, and KdInitSystem marked for INIT, this would have resulted in a crash...during a system crash, if /CRASHDEBUG had been enabled. Fun.
svn path=/trunk/; revision=55680
- Fix regressions which resulted in hang after reseting device
- Introduced in revision 55662
- Tested in VBox + Mass Storage Device
svn path=/trunk/; revision=55679
[CSRSRV/CSRSRV2]: Fix CsrLockProcessByPid -- it was skipping the root process itself, so requests coming from a CSRSS-hosted DLL would fail (which is exactly what server-to-server depends on).
[NTDLL]: Enable support for server-to-server calls!
Server-to-server calls work now, and one of the multiple mapping hacks is gone. User32 and kernel32 still need a bit of fixes to get rid of the other hacks, though. Also, Perf++.
svn path=/trunk/; revision=55677
- Fix heap code for 64 bit (before someone asks: the reason to use ULONG_PTR instead of ULONG is warning free code without casts)
svn path=/trunk/; revision=55675
- Fix error status of GetQueuedCompletionStatus, which should set the error to WAIT_TIMEOUT
- Fixes 4 failing winetests
- Patch by Samuel Serapion (samcharly_hotmail_com)
See issue #6907 for more details.
svn path=/trunk/; revision=55673
If the last chunk of the string sent to StringOut isn't a whole line and forcePrint is set to false, send back what's left after processing. The caller is then responsible for prepending that string next time it calls StringOut. Should fix the rest of debug log corruptions.
svn path=/trunk/; revision=55670
- Remove massive hacks from our port reset code
- We now wait for the reset complete interrupt instead of stalling 10ms and continuing
- Don't wait for a port to stabilize unless there's actually a device connected
[USBEHCI]
- Fix very long delay between setting the reset bit and clearing it
- Wait for the port to come back enabled before finishing the reset
- Don't wait for a port to stabilize unless there's actually a device connected
svn path=/trunk/; revision=55662
- Only send read format capacity when the device type is zero (direct access device)
- Add fixme
[INF]
- Fix issue
- USB CDROM installation now starts
svn path=/trunk/; revision=55661
- Implement timer routine which checks for hung srbs, not yet activated
- Stop timer when the device is stopped
- Implement function to check if the attached device is a floppy device
- Implement function to send read format capacity request
- Implement missing hardware id / device id support for floppy
- Add retry count parameter for debugging purposes
- Tested in VBox + VmWare + RH
svn path=/trunk/; revision=55659
- Don't send a bogus bus number (-1) to the PCI driver because then it tries to scan that bus and obviously doesn't find anything
- Legacy HAL can now enumerate boot critical PCI devices (like the USB controller) and build the device stack for USB boot
[HIVESYS]
- Missed this spot where usbhub is installed for an external hub causing a crash
svn path=/trunk/; revision=55652
[CSRSRV]: Enable some disabled code-paths and fix some locking patterns to be closer to what the code should look/be like. Correct ref count management is one of the things where regressions could show up later, so trying to nail these down now.
svn path=/trunk/; revision=55650
[CSRSRV]: Import server.c from CSRSRV2, and move the existing CsrSrv functions we had from wapi.c to server.c (with the reactOS-specific hacks we had). The other CsrSrv* functions are not yet used.
svn path=/trunk/; revision=55647
- First drop of a new implementation of NtAllocateVirtualMemory.
Very incomplete, don't hesitate to drop in and complete it.
For now, only MEM_RESERVE is kind of supported, as well as MEM_RESERVE|MEM_COMMIT
svn path=/trunk/; revision=55645
[CSRSRV]: Port from CSRSRV2 and use CsrApiPortInitialize instead of CsrpCreateListenPort. This will set appropriate SDs and also wait for all threads to be ready.
svn path=/trunk/; revision=55638
- Modify NtGdiSelectBitmap to correctly handle the case of pdc->dclevel.pSurface == 0
- Small code improvement without functional change for rtlstr functions
svn path=/trunk/; revision=55636
- Abort pipe when performing sync reset request
- add assert to check for bogus interface descriptors
- use endpoint max packet size when available
- flip data toggle after each transfer
- remove dead code
- use maximum of 4 pages or rest current buffer size when performing a bulk requests
- use nak reload count 3
- perform 1 transaction per frame
svn path=/trunk/; revision=55634
- use ARM3 allocation for PDE pages.
There is now one table for system PDEs, and their pages are not leaked anymore.
Also fix bug 5857 (aka mshtml bug).
Only for x86 now. Timo, I think you should do it for amd64 as well.
Thanks for your patience, go back to testing now.
A huge congrats to Cameron and Thomas, without them I wouldn't even have dared touching this code.
svn path=/trunk/; revision=55631
Use local labels in the inline asm code to avoid a clash when building with link-time code generation.
[PSEH]
Do not push memory operands with ESP as the base register (can happen when compiling with some GCC options), load arguments into registers.
Fix formatting.
svn path=/trunk/; revision=55628
[CSRSRV]: Port CsrSbApiRequestThread, CsrSbApiHandleConnectionRequest and all the Sb* APIs from CSRSRV2. Comment out parts that can't work yet.
[CSRSRV]; Port CsrSbApiPortInitialize, CsrSetProcessSecurity, CsrCreateLocalSystemSD, and almost all of CsrServerInitialization. Initialization is almost identical to CSRSRV2 now.
[CSRSRV]: Kill "Complete" routine support, nobody was using this.
svn path=/trunk/; revision=55615
[CSRSRV]: Flatten out the initialization code instead of a table of initialization functions.
[CSRSRV]: Make BasepFakeStaticServerData (our fake BaseSrv.DLL that's static inside CSRSRV for now) also create the BaseNamedObject Directory, along with the right ACLs and symbolic links, as well as the \Restricted sub-directory. Remove legacy function that was doing this before.
svn path=/trunk/; revision=55610
- Partially revert r54068. "My Video" is not a typo but name of registry value, which match Win2k3.
See issue #6568 for more details.
svn path=/trunk/; revision=55604
- Rewrite error handling
- Check if CSW is valid
- Check if the error handling was already started
- Reset device if required
- Error handling was completely broken and did not follow the reset procedure as defined in USB Mass Storage Specification Bulk Only Section 5.3.4
- Mass storage device now longer hang when receiving the read capacity request and ReactOS assigns a symbolic link
- Mass storage devices not yet fully working
svn path=/trunk/; revision=55601
- Remove bogus assert
- Store full configuration descriptor
- skip intermediate descriptors until the requested descriptor is found while parsing the configuration descriptor
- simplify CUSBDevice::GetConfigurationDescriptors
- check if an interface descriptor has endpoint descriptors
svn path=/trunk/; revision=55595
- If no control or one of wizard buttons has focus in PROPSHEET_SetWizButtons, set focus to default wizard button. Tested in Win2k3. Worse fix was already in diff but wasn't applied.
svn path=/trunk/; revision=55593
- Add support for LVS_EX_TRANSPARENTSHADOWTEXT style
[SHELL32]
- Add drop shadows support for desktop
- Set proper label color based on background color when drop shadows is disabled
- Update desktop colors when WM_SYSCOLORCHANGE or WM_SETTINGCHANGE msg is received
[BOOTDATA]
- Enable drop shadows by default (this option was hardcoded to 1 before)
[DESK]
- Send WM_SETTINGCHANGE with wParam = 0 when applying changes
[EXPLORER]
- Broadcast WM_SETTINGCHANGE message to shell view window
svn path=/trunk/; revision=55583
- Set endpoint direction indicator for control transfers
- Mask Version bits when printing out version
- Implement function for dumping endpoint descriptor
- Implement retrieving pid direction from setup packet
- Flip the pid direction for the status descriptor when data is transferred
- Return packet size 8 when using the control pipe
- OHCI control transfer hang is gone
- OHCI control should now work and attached HID devices finish initialization
svn path=/trunk/; revision=55574
- Fix code designed to avoid touching the hardware if it is already owned
[KDIO]
- Take ownership of the display if we're debugging to screen
[TXTSETUP.SIF]
- Add an example entry of DbgOsLoadOptions for debugging 1st stage to screen
- 1st stage setup can now be debugged completely to screen with a modified txtsetup.sif (comment 1st DbgOsLoadOptions and uncomment 2nd)
svn path=/trunk/; revision=55570
- Merge r55564 and r55567 from usb-bringup-trunk
- USB development should now be done in trunk exclusively (UHCI implementation may be an exception)
svn path=/trunk/; revision=55568
- Initilize ExDesktopObjectType->TypeInfo.ValidAccessMask and ExWindowStationObjectType->TypeInfo.ValidAccessMask to let win32k use access checks to desktop and window station objects
- When opening a desktop and window station during THREADINFO initialization, give full access to the objects
svn path=/trunk/; revision=55556
- We proudly merge the first charge of the usb-bringup branch. We do want to stress hardware support is still under heavy development and testing in real hardware is experimental
- Merge the Human Interface Device Stack(HID) which is used for mice / keyboards and other devices which use the USB interface, consisting of hidusb, hidparse, hidclass, mouhid, kbdhid
- Merge the composite driver, supports USB composite devices, laid out in usbccgp
- Merge the generic hub driver, which supports the USB root hub and in future USB hubs. Driver is usbhub
- Merge the Open Host Controller Interface driver (ohci)
- Merge the Enhanced Host Controller Interface driver (ehci)
- Merge the many fixes in other areas of ReactOS needed for USB to work (ntoskrnl, pci, inf, umpnpmgr, usetup)
- Special thanks goes the Haiku team, whose excellent code has provided a great base for the development of the new ReactOS USB / HID stack
- The development of the USB stack has shown the great potential when ReactOS developers team up together to achieve a common goal. The involved developers are here, listed alphabetically:
Alex Ionescu
Amine Khaldi
Cameron Gutman
Johannes Anderwald
Michel Martin
Thomas Faber
Thomas Lotz(Haiku)
Let's start the ReactOS revolution
svn path=/trunk/; revision=55555
- Enable async park mode when available
- Use correct type for the pipe handle
- Implement URB_FUNCTION_GET_STATUS_FROM_INTERFACE, URB_FUNCTION_GET_STATUS_FROM_ENDPOINT
- Verify that there is buffer provided in SubmitSetupPacket
[KBDCLASS]
- Forward requests to lower device
svn path=/branches/usb-bringup-trunk/; revision=55552
- In reactos when an application tries to use a desktop or window station object we never check if it should have access to the object. However the only exception to this is NtUserGetObjectInformation. Since desktops and windows stations are a mess, NtUserGetObjectInformation always fails. Ironically windows do such access checks everywhere except NtUserGetObjectInformation. Fix NtUserGetObjectInformation by removing access checks
- Note: access checks for every other place will be fixed soon
svn path=/trunk/; revision=55550
- Update pti->pClientInfo->hKL when pti->KeyboardLayout changes
- Fixes assertion after activating a second keyboard layout
svn path=/trunk/; revision=55549
- Implement retrieving endpoint halted status
- Fix multiple bugs which leaded to crashes (wrong device object passed). Usbstor now successfully restarts requests after handling the errors
- Implement mass storage reset, not yet used
- Add checks in CSW completion
svn path=/branches/usb-bringup-trunk/; revision=55548
- Fix incorrect maximum bound for resource requirements
- Handle PCI buses manually
[HAL]
- Don't hang if the IRP_MN_QUERY_INTERFACE is received
[INF]
- Misc fixes
svn path=/branches/usb-bringup-trunk/; revision=55541
- Ignore invalid extended capabilities pointer
- Don't hang on unimplemented IRPs that we actually use
svn path=/branches/usb-bringup-trunk/; revision=55539
- Implement IopFixupResourceListWithRequirements to replace IopCreateResourceListFromRequirements
- This functions fills in gaps between required resources and currently allocated resources in the resource list
- It also implements support for alternate requirement descriptors which IopCreateResourceListFromRequirements ignored
- Fix a critical bug in requirement processing that could result in the kernel allocating resources outside the devices requested range
- Now the PnP manager will actually claim resources that the device reports that it needs but it does not currently have
svn path=/branches/usb-bringup-trunk/; revision=55537
- Store configuration index in the setup packet
- Perform retrieving the configuration request before storing the configuration details. Perform the same action for the select interface request
- EHCI now supports selecting configuration & interface on devices with more than one configuration (previously only one configuration was supported)
- Code has not yet been tested, as no available usb device has such feature
svn path=/branches/usb-bringup-trunk/; revision=55536
- Don't corrupt memory when a queue head / transfer descriptor is freed
- How did this work before...
svn path=/branches/usb-bringup-trunk/; revision=55525
- Enable new heap code and cleanup the heap, releasing unused memory, before starting reactos
- Gives our kernel about 4MB additional memory, which - at the current leak rate - means around 8 minutes longer before it runs out of memory :)
- Using the temp heap for all temporary allocations should increase the amount of saved space
svn path=/trunk/; revision=55518
- Use the parse context to store whether or not the desktop object was created. If the object was not created, NtUserCreateDesktop should return immediately (this can happen if it didn't exist)
- Before this , if the desktop already exited, we opened the existing desktop and initialized it again. We also created another desktop heap, desktop window, etc.. This fact confused threads using this desktop and caused problems like the assertion we see in the test suite
svn path=/trunk/; revision=55517
- Ignore error when the device provides a string descriptor but no contents
- Fixes installation of cheap usb devices
svn path=/branches/usb-bringup-trunk/; revision=55514
Use MiNonPagedSystemSize when creating the system NP memory area instead of making assumptions about the memory layout
Finally the amd64 port boots into usermode and usetup is partly working, formats the harddisk but freezes, when copying files.
svn path=/trunk/; revision=55513
[SMSS]: Turn off all code other than setting up the pagefile.
There's a lot of debug prints, as soon as things seem stable after a few days, SMSS will be gone and SMSS2 will take over and DPRINT1s will mostly be gone.
svn path=/trunk/; revision=55509
Slightly reorganize architecture dependent headers, so that ASSERT can be used
[CRT]
Use some definitions only for vista+, since 2k3 msvcrt doesn't have them
svn path=/trunk/; revision=55504
- Add back old implementation for testing purposes (if'd out)
- Display more status field of queue head / transfer descriptor
- Dump queue head when a halted queue is detected
- Display data toggle status
svn path=/branches/usb-bringup-trunk/; revision=55502
[CSRSRV]: Fix many bugs in ServerSbApiPortThread. This function never worked as SMSS never actually connects to the SB API Port. Since SMSS2 does, a few bugs were discovered. Also, hack-plement the one SB API that is required for SMSS2 (SbpCreateSession). All we do is resume the thread we receive (Winlogon)... normally we'd also register a hard-error port, etc.
[CSRSRV]: If connecting with SMDLL to SMSS did not work (such as on my system where SMSS is no longer there), try using SMLIB and SMSS2 instead. This way CSRSS still works with both environments.
[CSRSRV]: Required rbuild/cmake changes to build with SMLIB.
svn path=/trunk/; revision=55500
- Don't wait for the work item to complete in the SCE completion handler
- Use an event to signal the completion of a reset and wait for that event during IRP_MN_START_DEVICE handling so devices are enumerated synchronously which allows USB boot devices to be enumerated
svn path=/branches/usb-bringup-trunk/; revision=55493
- Fix a few bulk transfer bugs
- Enable interrupt completion on last interrupt
- Increment transferred offset
- Mass storage device now gets further (passing SRB_FUNCTION_CLAIM_DEVICE)
svn path=/branches/usb-bringup-trunk/; revision=55491
- Rewrite the transfer descriptor handling
- Take into account the direction of the pid
- Revert previously added hacks
- Support now more than one transfer descriptor for control transfers
- Include the bulk transfers in one queue head which should speed up operation
- Store data toggle in usb endpoint structure
- Mass storage devices now passes the control status phase on real hw and enters bulk operation stage
svn path=/branches/usb-bringup-trunk/; revision=55489
Fix a FIXME in IopCreateArcNames.
This allows ntoskrnl to find where it boots from and where it is when performing remote boot.
Now ReactOS should boot a bit farther using PXE :-).
svn path=/trunk/; revision=55485
- Don't set NumberOfTransaction 0x3 for the async queue head(Haiku doesnt do it too)
- Also don't set maximum packet length or interrupt on complete
- Stop the async / periodic schedules after acquiring ownership from os (ported from the Pedigree OS)
- Remove unused structs from header
- Allocate device descriptor from start of a page to prevent alignment issues
- Clear halted bit in the queue overlay transaction layer (usb_queue.cpp:507)
- Always set the alternative and next pointers in BuildControlTransferQueueHead
- Use data the toggle in the transfer requests
- Start using toggle in bulk requests, invert data toggle after each descriptor
- EHCI controller still reports error 0xA / 0xB which is USB Error Interrupt + Frame List Rollover + [USB Interrupt]
- Errors need to be researched more, though the control transfers results appear o.k.
- Tested in real hw + ros stack with USB Mass Storage Device
- Currently hangs after first completed CDB/Data/CSW block
svn path=/branches/usb-bringup-trunk/; revision=55483
[SMSS2]: Implement the main LPC loop and handle all the supported APIs (all stubs for now). Also handle new connection requests and implement SmpHandleConnectionRequest.
[SMSS2]: Implement subsystem helper functions.
[SMSS2]: Use SmApiPort2 instead of Sm2ApiPort.
[SMSS2]: Rename the SMSRV_APIs not to conflict with the function names, nor with the client functions in smlib.
svn path=/trunk/; revision=55478
- Wait for periodic scheduling to start
- Increase the wait interval for asynchronous scheduling to start
svn path=/branches/usb-bringup-trunk/; revision=55477
- Implement MiIsUserP*e inline functions to efficiently determine if a pte is a pte/pde/ppe or pxe for a user mode page
- Make MiDetermineUserGlobalPteMask work for _MI_PAGING_LEVELS >= 3
- In MiResolveDemandZeroFault, use MI_MAKE_HARDWARE_PTE only, it does all the necessary work.
svn path=/trunk/; revision=55472
[SMSS2]: Implement SmpLoadSubsystemsForMuSession which correctly creates a session, loads win32k.sys inside of it, and then detaches from the session. Winlogon.exe is set as the initial command (but not launched), and CSRSS is fake-started.
[CSRSRV]: No longer start win32k.sys, that's not your job.
[SMSS2]: Kill most debug prints that aren't errors.
svn path=/trunk/; revision=55467
- Modify the logic in MmArmAccessFault, so that faults on kernel mode addresses including page table addresses are handled in the first part and user mode addresses (VADs) in the second part. This works, because for the special case of page table addresses, the user mode and kernel mode part of the code end up doing the same thing and this simplifies the code.
- In the user mode part call MiCheckVirtualAddress early and bail out if no VAD is found, since we do not care about any other cases any more.
svn path=/trunk/; revision=55459
- Cleanup Win32kProcessCallback and Win32kThreadCallback
- Even though it works, trying to post a message in Win32kProcessCallback after the THREADINFO is destroyed can only cause trouble, so move it in Win32kThreadCallback
- There is no need to try opening a desktop and windowstation before winlogon initializes
- Handle errors properly when we fail to get a default window station and desktop for the new process
- Enable win32k syscall hook callbacks
- Rewrite SetThreadDesktop to update THREADINFO properly and handle errors properly
- Do not initialize the THREADINFO in GetW32ThreadInfo as it is now done properly (actually leave a small hack that updates pci->dwTIFlags)
- Add UserDbgAssertThreadInfo that asserts the integrity of THREADINFO, CLIENTINFO and CLIENTTHREADINFO. This is called by GetW32ThreadInfo and the syscall hook callbacks
svn path=/trunk/; revision=55458
- Implement MiSynchronizeSystemPde, which does what its name suggests, synchronize a system PDE and is an improved replacement (with a more proper name) for MiCheckPdeForPagedPool
- Move some code to avoid an additional check
- Call MiResolveDemandZeroFault directy instead of creating a demand zero PDE and then calling MiDispatchFault, which after a lot of checks will finally do the same
svn path=/trunk/; revision=55456
- modify MiResolveDemandZeroFault to take the page protection as a parameter instead of the PTE. This will be used soon.
- Move increment of counters into the region where the pfn lock is held to avoid race conditions and make heavy locked operations unneccessary
- Add some ASSERTs
svn path=/trunk/; revision=55454
[SMSS]: Neuter away most of the code and only keep SMSS around for managing the API port.
Next step is to make the SMSS2 API port go into use.
svn path=/trunk/; revision=55453
- Add more structures to ntdddisk.h
- Fix more compile errors in disk_new
- disk_new builds but fails to link as classpnp import lib is broken
svn path=/branches/usb-bringup-trunk/; revision=55451
- Implement ResetController and call it before setting up the controller
- Don't disable 64-bit addressing until the BIOS gives up control of the controller
- Comment out manually clearing the SMI interrupt and the BIOS owned flag by request of janderwald
- EHCI no longer hangs for me on real hardware during controller initialization
svn path=/branches/usb-bringup-trunk/; revision=55450
[NTDLL]: Fix (although I'm not sure quite why) LdrpCheckForLoadedDll which was crashing now that I've fixed PAGE_EXECUTE sections with only FILE_EXECUTE handles. Ironically, this meant that LdrpCheckForLoadedDll never worked until my previous fix some revisions ago, it always returned FALSE.
This should fix KVM/QEMU crashes...
svn path=/trunk/; revision=55448
In PeFmtCreateSection don't use the ImageBase field of the ImageOptionalHeader directly, since it's different between PE32 and PE64. Use the ImageSectionObject->ImageBase field instead, which already contains the correct value.
svn path=/trunk/; revision=55446
- Add MM_HAL_VA_END and use it in MiAddHalIoMappings instead of making assumptions about the page tables
- Add MM_SHARED_USER_DATA_VA for x86
- use MM_SHARED_USER_DATA_VA in Mm to avoid ambiguities
- fix or comment out USER_SHARED_DATA in asm headers, as it should be the kernel mode address here
- set Teb->ExceptionList to NULL on amd64, it is used as a link to the Wow64 TEB, if any
svn path=/trunk/; revision=55445
- When rosautotest is launched, show how much time has past since the machine has started
- Also show how much time each individual test need to complete
- The purpose of this feature isn't to benchmark the os but to let us spot regressions regarding test time
svn path=/trunk/; revision=55441
- Check if the report uses a report id
- Fix arrows in USB 2 PS/2 conversion table (Needs more fixing)
svn path=/branches/usb-bringup-trunk/; revision=55440
- load gpsi->dtCaretBlink and g_PaintDesktopVersion in sysparams.c
- Rename IntGdiGetLanguageID to UserGetLanguageID
- Remove some unused functions and definitions
svn path=/trunk/; revision=55439
- Modify MiRosTakeOverPebTebRanges to only create a memory area for the shared user page and rename it to MiRosTakeOverSharedUserPage. Previously it was creating a memory area for the whole region from USER_SHARED_DATA up to MM_HIGHEST_USER_ADDRESS, which is the majority of the x64 user mode address space and doesn't even contain the PEB/TEB. Those are allocated below the shared user page and get their memory areas created in MiInsertNode.
- Add amd64 versions of MmGetPageTableForProcess, MmUnmapPageTable, MmGetPageFileMapping
svn path=/trunk/; revision=55438
- Rewrite SeCaptureSecurityDescriptor. The old code was mess and totally broken for 64 bit.
- Many fixes to security descriptor code that was making wrong assumptions about the SECURITY_DESCRIPTOR structures
svn path=/trunk/; revision=55437
- Check if attaching to device stack failed. In that case fail the request
[HIDUSB]
- Always set SupriseRemovalOk
[USBCCGP] [HIDCLASS] [USBHUB] [USBSTOR]
- Check if the pdo is still in the pdo list. Only call IoDeleteDevice in that case
svn path=/branches/usb-bringup-trunk/; revision=55429
- Bulgarian translation update of several modules (bug no. 6840), by CCTANEB, minor fixes by me;
- Polish translation update of sysdm by me;
svn path=/trunk/; revision=55424
- Fix a bug in CmpSetSystemValues, where an uninitialized handle would be closed in the failure path.
- Add a hack on top of the MI_GET_ROS_DATA(x) hack so that we can squeeze a pointer into a 32 bit field.
Make MmInitializeProcessAddressSpace amd64 ready and use a portable way to determine the page table base pfn in MiInitializeWorkingSetList
- Make MmProbeAndLockPages ready for 3 and 4 level page tables
add MiIsPteOnP*eBoundary macros - use these in MmProbeAndLockPages
- Raise IRQL to SYNCH_LEVEL not DISPATCH_LEVEL in KiAcquireDispatcherLock
- Add MiNonPagedSystemSize for all architectures
- Fix amd64 definition of KERNEL_HANDLE_FLAG
- Fix definition of PrototypePte
- Fix KiGetLinkedTrapFrame()
- Make MmProtectTpPteMask 64 bit wide
- Fix definition of MI_PTE_LOOKUP_NEEDED for amd64
- Impllement KiSendEOI() to be able to send an EOI from C code.
- Fix some MSVC/amd64 warnings
svn path=/trunk/; revision=55423
- Don't map the low 1 MB into kernel space, it's pointless and leads to inconsistencies between reported memory type and what is found in the page tables. On x86 it works, because it uses an overcomplicated pfn initialization wih redundent checks. Anyway this memory, containing freeldr and real mode code, was simply wasted.
- Silence some debug prints
- Fix warnings
svn path=/trunk/; revision=55421
- Define USBDI_VERSION to 0x500 instead 0x600 because we support none of the Vista APIs so USBD will report the correct version
svn path=/branches/usb-bringup-trunk/; revision=55420
- Add and fix missing exports
- Add UNIMPLEMENTED to unimplemented functions
- Fix completely broken USBD_GetPdoRegistryParameter (which wasn't even exported because somebody made a typo and couldn't figure it out)
- My Linksys USB300M driver now loads but it runs into some unimplemented code paths in usbehci
svn path=/branches/usb-bringup-trunk/; revision=55416
- Fix RtlLengthSecurityDescriptor
- Implement amd64 version of Implement RtlInitializeContext
- Add unwind info to amd64 debug asm functions
- Fix 64 bit HEAP_COMMON_ENTRY structure The resulting version doesn't exactly match the original windows one, but its compatible, as every field, except the dummy fields - which are omitted - is at its correct position.
svn path=/trunk/; revision=55415
- Implement amd64 version of LdrInitializeThunk and KiUserApcDispatcher
- Export ExpInterlockedPopEntrySList* on amd64, too
- Fix some 64 bit issues
svn path=/trunk/; revision=55414
- fix ASSERTs in amd64 inline functions
- fix definition of KWAIT_BLOCK
These fixes are not critical, so I don't update wdm.h to avoid additional recompilation
svn path=/trunk/; revision=55413
- Change LogfInitializeNew and LogfInitializeExisting to return an NTSTATUS code instead of a BOOL value.
- LogfInitializeExisting no longer calls LogfInitializeNew when reading a log file fails but returns a proper status code instead.
svn path=/trunk/; revision=55406
- Fix stack alignment in KiSwitchToBootStack
- Handle ExceptionFrame == NULL in KeContextToTrapFrame and KeTrapFrameToContext
- Implement KiSwapContextInternal
- Fix KiSwapContext and KiThreadStartup
- Implement dispatching of user mode exceptions including in-paging of module data used by the kernel-debugger
- Implement KeInitializeInterrupt, KeConnectInterrupt, KeSynchronizeExecution
- Don't zero more than the actual PCR size in KiInitializePcr
- Add asm function KiInitializeSegments to initialize the segment selectors to proper values
- Initialize system call entrypoints in KiInitializeCpu
- Implement KiDpcInterruptHandler, KiIdleLoop, KiInitializeUserApc, KiSwapProcess, KiSystemCallHandler, KiInitializeContextThread, KiSwapContextResume
- Implement asm functions KiRetireDpcList, KiInterruptDispatch, KiSystemCallEntry64, KiZwSystemService
svn path=/trunk/; revision=55405
- Add MI_REAL_SYSTEM_RANGE_START, which is the canonical system space base address, while MmSystemRangeStart variable contains a broken value, like on Windows.
- Define MmSystemRangeStart to MI_REAL_SYSTEM_RANGE_START, since on amd64 there is no dynamic system space start address and we don't want to use the broken address anywhere
- Add some more address layout definitions
- Add MiIsPteOnP*eBoundary() macro
- Add MMPPE, MI_WRITE_VALID_PPE and ValidKernelPpe definitions
- Fix initialization of PrototypePte
- Add mappings for VAD bitmap and working set list to MiInitializePageTable
- Fix calculations in MiBuildNonPagedPool
- Improve MiBuildSystemPteSpace
- Implement MiBuildPfnDatabase, MiAddDescriptorToDatabase, MiBuildPfnDatabaseFromPageTables, MiSetupPfnForPageTable all of these are written to be portable, yet they are untested on anything else than amd64 builds
- Mostly finish MiInitMachineDependent
svn path=/trunk/; revision=55403
- Add HID devices to the critical device database
- Fixes issues with mouse/keyboard not working until new device wizards are dismissed (USB drives work too)
[UMPNPMGR]
- Fix install failure after the kernel detects a critical device
[NTOSKRNL]
- Finally enable the proper IopResetDevice code
- Driver updating without a reboot for running devices is supported now (if someone will write the GUI)
[USBEHCI][USBOHCI]
- Fix removal bugs
svn path=/branches/usb-bringup-trunk/; revision=55401
- Rewrite MmFindGapBottomUp and MmFindGapTopDown, the old versions were broken and were first checking the address range after the first memory area and only used the area below (above) the first memory are when nothing free was found.
- Fix an ASSERT, that gets triggered now that the memory areas are created at the "right" locations
- Create a memory are for the boot loaded images, which previously could be overwritten happily by new memory areas, which was only prevented by the brokenness of the code
- Fix a few memory regions so that they are correct for amd64 builds as well
svn path=/trunk/; revision=55397
- Remove hack of forwarding pdo requests to lower device object of fdo
- Implement handling of reset port and cycle port
- IOCTL_INTERNAL_USB_RESET_PORT / IOCTL_INTERNAL_USB_CYCLE_PORT now need to be implemented in usbhub
svn path=/branches/usb-bringup-trunk/; revision=55396
- Properly handle an empty list in SmpSaveRegistryValue
- Compare the value string for equality, as the comment says, not inequality
- This fixes a crash, although another solution may perhaps be better
svn path=/trunk/; revision=55391
- Check if the transfer descriptors reported errors and propagate the error in the urb status field
- OHCI now reports error which the class driver are handling
- Allocate device descriptor from non paged pool to prevent possible alignment problems
- Fix checking of completed transfer descriptors
- Fix double freeing of descriptors
- Cleanup endpoints when the halted bit is set by the host controllers
- Check for the endpoint direction in the data descriptor
- Setup descriptor needs buffer rounding set
svn path=/branches/usb-bringup-trunk/; revision=55389
- Use constants instead of hardcoding register offsets. Taken from Haiku
- Elements in HCSP-PORTROUTE are 4 bit entries instead
svn path=/branches/usb-bringup-trunk/; revision=55385
- Write the partition table back after CheckActiveBootPartition() because it could've changed the partition table on a disk with no active partition
- Fixes issues installing ROS on new disks and disks with no active partitions (including USB disks)
svn path=/branches/usb-bringup-trunk/; revision=55382
- In 2012 a long protected natural resource for bugs was shutdown due to popular usb device support demand
- Fix length check in USBD_ParseDescriptors. (PLONG)Descriptor + TotalLength is not the length of the configuration descriptor but twice the configuration descriptor
- Reset found variable in the loop. This fixes finding of interfaces which are not the first one in the configuration descriptor
- Again moving to next usb descriptor was broken. UsbDeviceDescriptor = UsbDeviceDescriptor + UsbDeviceDescriptor->bLength does not move to the next but to next XX descriprors specified by bLength
- Introduced in rev 17382 (14/08/2005)
svn path=/branches/usb-bringup-trunk/; revision=55370
- Wait 100 milliseconds for the port power and insertion to stabilize
- Wait for the 10 millisecond "recovery" period after reset
svn path=/branches/usb-bringup-trunk/; revision=55367
- Use report value when report item bitcount > 1
[KBDHID]
- Print raw report
- KBDHID starts to receive normal keys and modifier keys
svn path=/branches/usb-bringup-trunk/; revision=55364
- Attempt to fix the font leak when switching applications. Tested with other applications other than Acrobat Reader since it crashes when installing. See bug 5314.
svn path=/trunk/; revision=55362
- Fix bugs in HandleClassEndpoint, HandleClassInterface which always set direction device to host instead of relying to the transfer flags provided
- Allow setup packets w/o a data stage
svn path=/branches/usb-bringup-trunk/; revision=55360
- Implement HidP_GetUsagesEx
- HidP_GetUsagesEx uses HidP_GetUsages with undefined usage page
- Required for hid usb support
- Check if there is a maximum set. In that case verify if a usage maximum has been set and apply it to the currnt usage. Fixes asserts while usb composite device installation (keyboard)
svn path=/branches/usb-bringup-trunk/; revision=55357
- Implement set idle request. The device should only send data when there is actual data. May also fix the mouse debug flood when mouse is ejected
- Not yet supported in OHCI
- Add missing constants
svn path=/branches/usb-bringup-trunk/; revision=55356
- Don't check for the correct parent in IopActionInitChildServices
- This function is also used to load drivers later for devices enumerated early in boot (before we have access to the disk) and we want it to run for all devices on the system, not just those that are immediate children of the root node
- Fixes broken PnP in 3rd stage
svn path=/branches/usb-bringup-trunk/; revision=55355
Implement IoVolumeDeviceToDosName(), taking advantage of newcomer in the branch, MountMgr :-).
This implementation will in fact don't work as MountMgr doesn't implement that part... But the day it's ready, this function will be!
svn path=/branches/usb-bringup-trunk/; revision=55354
- Rewrite hidparser so that it support multiple top level collections
- Add support for flat collection context meaning all context structures are aligned in after each other. The reason this is required is that usermode hid.dll will access the same structure and thus can't use kernel pointers. User mode hid.dll is used to support keyboard with custom keys (for example search key, email key etc)
svn path=/branches/usb-bringup-trunk/; revision=55353
- Missed a fix from last commit
- Wait for the device status change work item to complete
- The USB stack is now properly constructed for USB booting
svn path=/branches/usb-bringup-trunk/; revision=55339
- Start support for multi top level hid collections
- Not yet full implemented
- Required for proper usb composite support
svn path=/branches/usb-bringup-trunk/; revision=55332
- Create PDO earlier so ACPI can be loaded earlier in boot
[HIVESYS]
- Create critical device database with required devices for USB boot
[NTOSKRNL]
- Read the CDDB and install devices that are critical for boot
- This implementation won't be 100% correct until a PnP manager rewrite but it should work well enough for our needs
- Not complete and disabled atm
svn path=/branches/usb-bringup-trunk/; revision=55328
- Add function to append interface number
- Append function number to BusQueryDeviceId / BusQueryInstanceId
- USBCCGP is now properly enumerated and its child device start to get installed
svn path=/branches/usb-bringup-trunk/; revision=55326
- Fix bug while scanning configuration descriptor
- Store interface list in the PDO device extension
- Display interface numbers when dumping function descriptor
- Fix a few bugs while selecting the configuration for the PDO
svn path=/branches/usb-bringup-trunk/; revision=55325
- Pass the interface number in the request when issuing hid get report descriptor
- Fixes wrong report descriptors being obtained
svn path=/branches/usb-bringup-trunk/; revision=55322
[SMSS2]: Implement all the required registry configuration parsing functions. KnownDLLs, DosDevices, PageFiles, Subsystems, ExcludedKnownDLLs, PendingFileRenameOperations, ObjectDirectories, Environment keys are all processed and put into lists. Other than creating the object directories and setting the environment variables, though, no other work happens -- but we do DPRINT1 what should happen.
svn path=/trunk/; revision=55319
1. Belief that Ob is perfect.
Fix a major bug which allowed user-mode applications to get kernel handles if they so requested -- of course, they woudl then be completely unable to use such handles. This pattern is seen in Rtl code which we share, and where the intent is to give kernel kernel handles, and user user handles, so OBJ_KERNEL_HANDLE is unconditionally used, which in NT had the right effect. For us though, it gave unusable user-handles. This had the direct effect of completely breaking RtlQueryRegistryValues from user-mode if the TOPKEY or SUBKEY flags were used.
svn path=/trunk/; revision=55318
- Fix selection of the active partition
- Running 1st stage setup to a USB drive now results in a bootable USB drive that successfully runs FreeLoader which can load the ROS kernel, HAL, and other critical boot executables from the USB drive but ROS fails early in boot after FreeLoader hands control to the kernel
svn path=/branches/usb-bringup-trunk/; revision=55316
- Handle SRB_FUNCTION_SHUTDOWN successfully (failing it results in a stack overflow due to a bug in disk.sys)
- Fixes the crash during shutdown after using a USB drive
svn path=/branches/usb-bringup-trunk/; revision=55315
- Formatting fixes
[TXTSETUP.SIF]
- Boot load certain USB drivers in 1st stage for eventual support for installing from USB
svn path=/branches/usb-bringup-trunk/; revision=55309
- Fix the ordering of disks in the 1st stage setup list
- Search for the active partition on the selected install disk instead of the first disk
svn path=/branches/usb-bringup-trunk/; revision=55299
- Enable device as pci bus master
- Fixes hanging of Vmware USB EHCI controller
- Revert changes from 55293 from usb_queue
- Currently crashes when usbstor wants to receive descriptors
svn path=/branches/usb-bringup-trunk/; revision=55294
- Update async address everytime a new queue head is added as specified in EHCI spec section 3.2
- Add more checks
- Vmware EHCI not yet working
svn path=/branches/usb-bringup-trunk/; revision=55293
- Log files which are opened when the eventlog service is started are permanent files and will only be closed when the service is stopped. Other log files are closed when their handle is closed.
- Opening a non existing backup log file must fail.
- Opening an existing backup log files opens the file for reading only.
svn path=/trunk/; revision=55279
* Import cor.idl, cordebug.idl and metahost.idl from Wine.
* Remove the now unneeded cor.h.
* Update corerror.h, corhdr.h, mscoree.idl and winsxs.idl from Wine.
* Update the cmake and rbuild files to reflect the changes.
svn path=/trunk/; revision=55272
- Implement counting of all interface descriptors
- Store device descriptor and configuration descriptor in pdo device extension
- Store all interface descriptors for an composite audio device
- Implement parsing and constructing a special configuration descriptor for each individual device function
- Implement URB_FUNCTION_GET_DESCRIPTOR_FROM_DEVICE for device descriptor and configuration descriptor
- USBGCCGP now receives select configuration request
- Tested in XP + Ros USB Stack + USB Audio Device
svn path=/branches/usb-bringup-trunk/; revision=55261
- Use Function index than interface number. Fixes invalid hardware ids
- Fix construction of compatible id
- USB Audio device now installs on XP when usbccgp is used
- Now asserts in sb-bringup-trunk\drivers\usb\usbccgp\pdo.c:343) PDO_Dispatch Function f not implemented, which is implemented next
svn path=/branches/usb-bringup-trunk/; revision=55255
- Partial sync to wine 1.3.37 without the ComCtl32 diff applied. This fixes the hangs and the exceptions during testing. Researching DrawFocusRect issues. Please Developers, note any information for the diff other than what is posted on svn, also this is open to anyone to be synced.
svn path=/trunk/; revision=55254
- Implement counting interface descriptors
- Fix allocating interface list
- Fix multiple bugs when copying interface information
- Implement support function to dump function descriptor
- Fix multiple bugs when creating compatible / hardware id string
- Implement legacy enumerator, which will be used for HID composite devices
- Implement special enumerator for audio devices
- Display static GenericCompositeUSBDeviceString, needs to be read from registry in order to be MUI aware
- Device initializes and device wizard pops up
- Tested in XP with Ros USB stack + USB Audio Microphone
svn path=/branches/usb-bringup-trunk/; revision=55253
- Rewrite GetDeviceIds function
- Don't rely on hardcoded constants when building the device id strings
- Don't corrupt the device id string when building the instance id string
- Fix bug in GetUsbStringDescriptor which read beyond the allocated string when copying the result.
svn path=/branches/usb-bringup-trunk/; revision=55238
- Fix initialization bugs for EHCI controllers
- Try again to release ownership of low-speed devices after reset
- Wait for the port reset to complete
svn path=/branches/usb-bringup-trunk/; revision=55232
For your unique, viewing and executing pleasure..
For one time only....by popular demand....
The one...
The only...
The unforgettable, unforgivable...
Meanest, baddest, piece of code around...
He's back..... with a vengeance...
The ultimate...
Mega...
Uber..
Amazing...
HACK
OF
DOOOOOOOOOOOOOOOOOOOOOOOOOOOOOM
</applause>
svn path=/trunk/; revision=55227
- Set the correct transfer direction value in the CBW
- Fixes the infamous system hang that occurred when writing to a USB drive
svn path=/branches/usb-bringup-trunk/; revision=55223
- Implement querying device relations for FDO
- Implement creating device objects for each discovered function
- Invalidate device relations after the child pdo have been created
- Implement PDO device relations, most of query id, query device capabilities, start device
- USBCCGP should now initialize and create the child pdos
- Next is device control handling
- WIP, untested
svn path=/branches/usb-bringup-trunk/; revision=55219
- Merge r55212 and 55215
- Add a temporary hack to allow testing installation on a device not bootable by the BIOS (like USB drives in VirtualBox)
- USB devices now show up as installation targets during setup but attempting to install onto them results in setup hanging due to bugs in the USB code
svn path=/branches/usb-bringup-trunk/; revision=55217
- Allow removable media to be selected as an installation target
- Fix broken unpartitioned space calculation if the first partition starts below the disk's track size (such as a 1MB aligned partition)
svn path=/trunk/; revision=55215
- Fix an off-by-one bug in the drive letter assignment code that prevented the 4th partition on a disk from being assigned a drive letter
svn path=/trunk/; revision=55212
- Query bus interface and check if USBC_DEVICE_CONFIGURATION_INTERFACE_V1 is supported. This interface is implemented by attached usb filter drivers and is used to enumerate functions of the composite usb device
- Implement enumeration of function by using usb interface association descriptors and by the USBC_DEVICE_CONFIGURATION_INTERFACE_V1
- Needs audio legacy method and union function descriptors to be fully functional
- WIP, not yet tested
- Fix build with mingw by declaring _DISK_GEOMETRY_EX outside the function
svn path=/branches/usb-bringup-trunk/; revision=55197
- Start implement USB Composite generic parent driver which is used for USB composite devices
- Start implement FDO initialization, needs to implement parsing of usb interface assocaition descriptor to complete FDO initialization
svn path=/branches/usb-bringup-trunk/; revision=55195
- Implement IRP_MN_REMOVE_DEVICE for FDOs and PDOs
- USB mice will now work after being disconnected then reconnected
svn path=/branches/usb-bringup-trunk/; revision=55194
- Multiple the scroll value by WHEEL_DELTA (120) like i8042prt does
- Scrolling with USB mice works now
- Disable debugging
svn path=/branches/usb-bringup-trunk/; revision=55191
- Properly dismiss the enable status changed interrupt to avoid an interrupt storm after a device is disconnected
svn path=/branches/usb-bringup-trunk/; revision=55189
- Fix GetPortStatus() and remove the cached status stuff (except for reset that we have to cache because the EHCI spec has no port reset complete bit)
svn path=/branches/usb-bringup-trunk/; revision=55188
- Revert r55167 now that OHCI is fixed
- USB drives attached to an OHCI controller before boot are now enumerated without a reconnect needed
svn path=/branches/usb-bringup-trunk/; revision=55187
- Remove the hacky way of determining if a device connect occurred (prone to all sorts of race conditions) and just always create a device since the only reason we reset right now is for a device connect
svn path=/branches/usb-bringup-trunk/; revision=55186
- Fix and enable the proper GetPortStatus implementation
- Remove the old hacked cached port status mess
svn path=/branches/usb-bringup-trunk/; revision=55185
- Create DEVICEMAP\VIDEO key with proper length. Fixes a bug exposed as "Arwinss unable to change resolution".
See issue #6838 for more details.
svn path=/trunk/; revision=55183
- Fix StartController() to perform initialization according to OHCI spec
- Fixes the infamous OHCI initialization hang on real hardware
svn path=/branches/usb-bringup-trunk/; revision=55181
- No need to turn off interrupts
- Check if bios is active
- Check for timeouts when resetting host controller
svn path=/branches/usb-bringup-trunk/; revision=55171
- Don't turn off interrupts before setting the OHCI_OWNERSHIP_CHANGE_REQUEST bit because it prevents the SMM driver from receiving the interrupt that tells it to give up ownership of the host controller
- This fix should be merged to Haiku also which has the same bug
svn path=/branches/usb-bringup-trunk/; revision=55170
- Add debug trace
[USBOHCI]
- Implement proper GetPortStatus
[USBHUB]
- Reset all connected ports before sending first SCE
- USB Devices present before booting are now detected with OHCI controller. EHCI code is present but not yet activated
svn path=/branches/usb-bringup-trunk/; revision=55167
- Rewrite pending SRB handling and change some behavior of the IRP queue
- The caller is no longer responsible for checking whether it can call USBSTOR_QueueNextRequest; frozen IRP queue and pending SRB are both handled for them
- It's no longer required for the caller of USBSTOR_QueueTerminateRequest to know whether the SRB was active (which was impossible before when handling a cancellation)
- Many potential race issues with IRP cancellation are eliminated
- Debugging hung SRBs is much easier now that pointer to the active one is stored in the FDO
svn path=/branches/usb-bringup-trunk/; revision=55157
Finally add the long awaited Mount Point Manager (aka mountmgr).
It's not complete yet, but will do most of the job it's been designed for.
What's missing is some dos volumes handlers, complete database support (local - registry - is complete, remote - files - is not).
It handles NTFS properties like reparse points.
It also handles devices with drive letters and those without drive letters (by using their unique ID).
Devices removal/arrival is detected through notifications (might be an issue).
Some work will have to be done in storage stack to fully support it.
Most of its IOCTL have been implemented (it's possible, for example, to notify a volume arrival through them, in case notifications don't work).
There's still some work to do on it to have it complete and fully implemented.
Enjoy!
svn path=/branches/usb-bringup-trunk/; revision=55156
- Fix bugs introduced in 55134, 55135
- USB Mass Storage devices should now automatically install again
svn path=/branches/usb-bringup-trunk/; revision=55147
- Fix cancellation for IRPs that have already been dispatched for processing by IoStartNextPacket
- Don't complete IRPs with the IRP list lock held
- Clear the cancel routine before completing the IRP
svn path=/branches/usb-bringup-trunk/; revision=55138
- Check if the device is a composite device
- Report USB\COMPOSITE as compatible id when a usb compsite device is detected
svn path=/branches/usb-bringup-trunk/; revision=55134
- Disable VS11 warning about cast between different function classes (IO_DPC_ROUTINE to KDEFERRED_ROUTINE) where needed
svn path=/trunk/; revision=55129
- Fix an if condition in VfatCreateFile causing opening a volume to always return access denied
- Fix some DPRINT issues
- Powered by VS11
svn path=/trunk/; revision=55127
- Use the same lock in the IUSBQueue as in the IDMAMemoryManager
- add debug traces (default off)
svn path=/branches/usb-bringup-trunk/; revision=55110
- Compute the frame interval correctly
- Fixes a deadlock on real hardware after enabling interrupts
svn path=/branches/usb-bringup-trunk/; revision=55094
- ReadProcessMemory/WriteProcessMemory only write to *lpNumberOfBytesRead/Written if user-mode passed in the parameter, as its an optional argument in Win32, but not in NT. Instead, use a local variable. This means that anyone calling ReadProcessMemory/WriteProcessMemory in ReactOS before with a NULL output argument (totally valid) was getting an error before!
- WriteProcessMemory actually returns STATUS_ACCESS_VIOLATION in a few cases, even if it's defined as a BOOL function. Code on Google shows major applications depending on this, which we weren't doing.
- Rewrite InitCommandLines to be much simpler. No normalization or copying or ANSi/OEM logic is needed.
- GetProcessAffinityMask should use the BaseStaticServerData from CSRSS instead of querying system information each time.
- GetProcessShutdownParameters and SetProcessShutdownParameters should use the LPC status code from CSRSS, not the LPC API status code.
- GetProcessWorkingSetSize now calls GetProcessWorkingSetSizeEx.
- Implement GetProcessWorkingSetSizeEx.
- SetProcessWorkingSetSize now calls SetProcessWorkingSetSizeEx.
- Implement SetProcessWorkingSetSizeEx.
- Acquire the required privilege in SetProcessWorkingSetSize(Ex).
- Fail with correct status code in SetProcessWorkingSetSize(Ex).
- GetExitCodeProcess should check if this is a VDM process and get the exit code that way.
- GetStartupInfoW should not fail if the input is NULL. It should crash.
- GetStartupInfoW was not filling out the lpReserved field, which should contain the ShellInfo buffer.
- GetStartupInfoW was always setting standard handles -- it should not do so if those are console handles.
- GetStartupInfoA was not thread-safe.
- GetStartupInfoA was assuming all Unicode->ANSI conversions will be successful.
- GetStartupInfoA was not filling out lpReserved either.
- ExitProcess was not using SEH and was not using the PEB lock.
- TerminateProcess was not setting ERROR_INVALID_HANDLE last error code.
- FatalAppExitA was not using static TEB buffer, and was always assuming success.
- FatalAppExitW was doing some sort of bizarre hack. It now raises as a hard error as it should.
- FatalExit now displays a debugger input interface on checked builds, just like Windows.
- SetPriorityClass now tries to acquire the real time privilege when needed, and handles failure to do so.
- GetProcessVersion rewritten to be cleaner and simpler.
- Annotate and reformat functions where needed.
- Rename lpfnGlobalRegisterWaitForInputIdle to UserWaitForInputIdleRoutine
- GetProcessPriorityBoost is now BOOL-safe.
- IsWow64Process now sets NT error code using only one API.
- CommandLineStringA/W -> BaseAnsiCommandLine/BaseUnicodeCommandLine.
svn path=/trunk/; revision=55092
- Don't send removal IRPs to device nodes already pending removal
- USB device removal and reinsertion works now
svn path=/branches/usb-bringup-trunk/; revision=55087
- Fix a broken check that resulted in freeing the same device object twice
- Enable the IoDetachDevice call in usbstor now that the kernel bug is fixed
svn path=/branches/usb-bringup-trunk/; revision=55086
- Fix IntReadConsoleOutputCharacter to copy the correct count of characters. Its 3rd parameter is a character count and not buffer size.
- Should fix infinite loop when kernel32:console test runs
svn path=/trunk/; revision=55081
- Implement device disconnect indication for usbehci and usbohci
- Implement device removal for FDOs and PDOs in usbstor and usbhub
svn path=/branches/usb-bringup-trunk/; revision=55080
- Load the disk class for USB drives in 1st stage
- They don't show up on the partition page due to some usetup stuff that needs to be changed
svn path=/branches/usb-bringup-trunk/; revision=55076
- Prevent an infinite loop in CdfsDeviceIoControl if the the verify request is successful but IoCallDriver keeps returning STATUS_VERIFY_REQUIRED
svn path=/trunk/; revision=55057
- Add a class2 hack to recognize and assign drive letters to storage devices added while the system is running
svn path=/branches/usb-bringup-trunk/; revision=55056
- Add Microsoft Mice Device Description
- fixes hid device entries where instead of device description the device id is displayed
svn path=/branches/usb-bringup-trunk/; revision=55054
- Add usbd to bootcd because usbohci depends on it
- Make device interface creation failure a non-fatal error so the port drivers start successfully in 1st stage
svn path=/branches/usb-bringup-trunk/; revision=55034
- Create a branch to do a proper merge of USB work from a trunk base instead of from cmake-bringup
- In the future, DO NOT under any circumstances branch another branch. This leads to merge problems!
svn path=/branches/usb-bringup-trunk/; revision=55018
- Implement IrqlToSoftVector() macro and use it in HalRequestSoftwareInterrupt, so that on amd64 the correct interrupt is triggered.
- Really start at the lowest vector for the IRQL in HalpAllocateSystemInterrupt
- Small code refactoring
svn path=/trunk/; revision=55014
- Add support for user name settings.
- Obtain user name / domain hints from registry.
- Pass user/domain info to the server if available.
- Add Czech translation.
svn path=/trunk/; revision=55008
- Use macros for calling IBanneredBar methods
- Remove some definitions from todo header because they are already in shlguid_undoc.h
svn path=/trunk/; revision=55004
- Fix some corner cases for settings parsing (empty values, values containing delimiters).
- If provided, take port number into account.
svn path=/trunk/; revision=55003
* Fix some memory leaks.
* Add some comments.
* Fix broken PcNewResourceSublist which ignored the MaximumEntries parameter.
* Add more checks.
svn path=/trunk/; revision=55002
- ReactOS now supports open and WEP encrypted networks (both ad-hoc and infrastructure) (WPA(2) will be supported later)
- Use the wlanconf tool to configure the WLAN adapter (after installing the XP driver for your WLAN adapter)
- "wlanconf -c <SSID>" connects to an infrastructure network. Adding the "-a" option connects to (or creates) an ad-hoc network. Adding "-w <WEP>" will enable WEP encryption using the supplied key.
- "wlanconf -s" will scan and display a list of the surrounding networks and various attributes like signal strength, SSID, BSSID, network mode, and supported rates
- "wlanconf" will display the current WLAN configuration details if the adapter is connected
- The DHCP service will detect network changes and refresh its state accordingly when associating with a new wireless network
svn path=/trunk/; revision=55000
- Give numeric identifiers to all dialogs
- Add file copy dialog to languages files where it was lacking
- Remove unused admin tools icon duplicate
svn path=/trunk/; revision=54997
- Avoid reporting a cryptic error code when wlanconf is used without any WLAN adapters present
- Print the BSSIDs when doing a scan
- Fix the WLAN disconnected check
svn path=/branches/wlan-bringup/; revision=54995
- Don't use shell32 functions when processing DLL_PROCESS_ATTACH because shell32.dll can be initialized later. Fixes shell32+wininet debug spam at boot.
svn path=/trunk/; revision=54992
- Deallocate the process page directory when destroying its address space (removed in r48233 and now resurrected in a version compatible with ARM3)
- Fixes leaking system pages on each process exit (868 MC_SYSTEM pages were allocated just sitting at the desktop on livecd without the patch, only 187 with the patch)
svn path=/trunk/; revision=54988
- Load shell extensions from Directory key only for file system directories. Fixes New menu being added to special folders context menu (for example My Computer or Control Panel)
- Load shell extensions from AllFilesystemObjects only for files and directories
- Hardcode special folders attributes in proper place
svn path=/trunk/; revision=54985
* Fix a bug which closed an already invalid handle.
* Allow only general registry keys to be deleted.
* Do not close a key twice in the error case.
svn path=/trunk/; revision=54982
- Don't load shell extensions from * key for background context menu. Fixes#6668 (background context menu not working after installing Notepad++)
- Properly set pidlFolder argument in extensions Initialize function call
- Fix openwithmenu.rgs file creating two keys instead of one
svn path=/trunk/; revision=54977
[KERNEL32]: Implement BasepSxsCloseHandles, BasepReplaceProcessThreadTokens, BasepIsProcessAllowed, BasepCheckWebBladeHashes, BasepIsImageVersionOk which will be needed for the future CreateProcess re-implementation. These functions partly support SxS, Safer (Authz) and Application Certification features added in XP/2003. We also emulate support for Computer Server, Web Blade Server and Embedded ReactOS. The last function does correct image version checks to prevent invalid binaries from running.
[KERNEL32]: Implement BaseUpdateVDMEntry and BaseCheckForVDM using the new CSRSS messages (not implemented on the server-side yet). Stubplement BaseCheckVDM. These will be needed for future VDM-stub-support (primarly so we can run 16-bit installers).
[KERNEL32]: Implement BasepFreeAppCompatData, BasepCheckBadapp, and IsShimInfrastructureDisabled (exported as BaseIsAppcompatInfrastructureDisabled). These stub most of the required/exported application compatibility APIs, as long as someone sets DisableAppCompat in \\Registry\\MACHINE\\System\\CurrentControlSet\\Control\\Session Manager\\AppCompatibility.
svn path=/trunk/; revision=54975
- Add New menu to context menu handlers. Fixes regression introduced by r54959.
- Remove unwanted files
See issue #6823 for more details.
svn path=/trunk/; revision=54974
- Fix GetRosInstallCD - GetLogicalDriveStringsW returns drive paths, not letters.
- When building intl.cpl command line for unattended setup don't use %S - swprint interprets it as multi-byte string. Use "%s" instead. This bug caused yesterdays KVM problems.
svn path=/trunk/; revision=54971
* Initializes object attributes with correct handle attributes.
* Use the provided access rights and create options.
See issue #5934 for more details.
svn path=/trunk/; revision=54969
[NTDLL]: Use HANDLE instead of Win32 HKEY.
[NDK]: Add missing NtCreateProcessEx flags and some missing Ldr functions + missing RtlGetFullPathName_UstrEx.
[CSRSS]: Define two new CSRSS messages (not implemented): UpdateVdmEntry and GetVdmExitCode.
svn path=/trunk/; revision=54968
- The DHCP service now recognizes adapter configuration and media state changes without manual intervention
- Multiple bugs with a static IP configuration are fixed
- IpRenewAddress (ipconfig /renew) and IpReleaseAddress (ipconfig /release) no longer corrupt internal DHCP state
- Routes that cannot currently be resolved by ARP no longer become corrupted by the neighbor timeout
svn path=/trunk/; revision=54966
- If there is no default menu item, make Open With menu item the default, rename it to Open and disallow it to be submenu.
See issue #6801 for more details.
svn path=/trunk/; revision=54964
* Improve the way we set the debugging flags for debug and release build types.
* Don't generate debug info when we're in prefast mode and don't prefix with prefast when linking executables.
* We don't need to alter the c/c++ compiler variables in prefast mode anymore.
svn path=/trunk/; revision=54962
- Rewrite Control_DoLaunch. Now if cpl has only one applet, applet name/id is ignored
- When launching control panel applets always execute rundll32 instead of using static handlers from registry
svn path=/trunk/; revision=54960
* Apply a default date of 00/00/0000 when DriverVer is missing. Should fix installation of drivers which do not provide a DriverVer directive.
See issue #6676 for more details.
svn path=/trunk/; revision=54958
- Move folder properties to CFileDefExt class and share as much code as possible. Now changing file attributes is implemented.
- Fix a warning (patch by EDIJS - #6821)
- Load property sheet handlers from "Directory" key
svn path=/trunk/; revision=54948
- Adapt the reconnect hack to also enable DHCP when transitioning from static configuration to DHCP enabled via netcfgx
svn path=/branches/wlan-bringup/; revision=54945
- Queue a work item from the ProtocolStatus handler so we can refresh the state from the registry instead relying on hacks to track our state
svn path=/branches/wlan-bringup/; revision=54944
- Store the NTEContext in the IP address list so netcfgx can remove IP addresses
- Fixes routing after manually setting an IP address
svn path=/branches/wlan-bringup/; revision=54942
- Configure route NCEs to never timeout to prevent them from falling out from under the routing code
- Fix a typo
- Fix detection of duplicate routes
[NETCFGX]
- Don't delete all routes for an interface when adding a default gateway
svn path=/branches/wlan-bringup/; revision=54939
- Display free and used space and capacity in bytes and gigabytes for all drives in drive properties dialog
- Format date and time in file properties according to system locale
- Initialize checkboxes in file properties dialog
- Improve polish translation
svn path=/trunk/; revision=54936
- Fix an issue assigning a private address after the IpReleaseAddress API was used prior to IpRenewAddress
- Move the IP refresh hack into dhcpcsvc so it now properly refresh IP information after an adapter is disconnected and reconnected (wired and wireless)
- When a state change occurs (connecting to a different wireless network or unplugging and plugging in the Ethernet cable), TCP/IP will flush routes and the neighbor cache then set the IP address to 0.0.0.0. DHCP will detect that IP address (that part is the hack since we do it via polling instead of events) then send a DHCP discover packet out via the new network connection. No more ipconfig /renew to get a new DHCP lease after network changes. It's all seamless now :D
svn path=/branches/wlan-bringup/; revision=54933
- Add "Description" string to resources. Don't show application file extension in Description field (for applications without version info). Patch by Edijs Kolesnikovis (#6819)
- Implement associated application icon loading in File Properties dialog
- Implement "Change" button in File Properties dialog
svn path=/trunk/; revision=54927
- Fix first chance exception if application has no strings in version info
- If Lang is not set use US English version info. Fixes WinRAR version info.
- Applications has Description filed instead of Opens With
- Load language in file version info
svn path=/trunk/; revision=54925
- Fix copying files in New menu. Fixes WinRAR new items.
- Don't show message if copying file failed because empty file is already created
svn path=/trunk/; revision=54923
- Simplify BITMAP_CopyBitmap, by using the SURFOBJ members directly instead of calling BITMAP_GetObject.
- Use RtlCopyMemory directly in BITMAP_CopyBitmap and remove IntSetBitmapBits
- Improve prototype of GreCreateBitmap(Ex) by using ULONG instead of INT for the bitmap dimensions (negative values are invalid)
- remove a duplicated type definition
svn path=/trunk/; revision=54922
- Release the current IP address before disconnecting from an AP
- Refresh the IP information after connecting to an AP
svn path=/branches/wlan-bringup/; revision=54914
- Delete the existing default gateway when releasing DHCP IP or transitioning to static IP
- Fixes bug #6205
svn path=/branches/wlan-bringup/; revision=54913
- Fix ipconfig hang with ipconfig /renew and ipconfig /release
- Fix duplicate protocol entries being added for the same adapter
- Use WSAEventSelect and WaitForMultipleObjects to manage waiting between timeouts, adapter state changes, and incoming packets
svn path=/branches/wlan-bringup/; revision=54912
- Move drive default property sheet pages to CDrvDefExt shell extension
- Read cleanup mgr path from registry
- Fix default extension being released in wrong moment
svn path=/trunk/; revision=54910
- Don't check exe files for Open With application
- If application is invalid display proper text
- Properly handle application parameters and quotes when parsing application command. Patch by EDIJS, improved by me.
See issue #6770 for more details.
svn path=/trunk/; revision=54906
- Simplify and fix getting file type description and icon in File Properties dialog
- Apply EDIJS's patch for scrolling text in location field on File Properties dialog (#6807)
svn path=/trunk/; revision=54905
- Make File Properties code safer and better
- Change version information when it user selects item in listbox (in previous version user had to double click)
- Show filename only in Open With dialog. Part of patch by EDIJS (#6772)
svn path=/trunk/; revision=54904
- Store the correct length in Irp->IoStatus.Information so bytes don't get chopped off
- Fix multiple bugs in the receive path (which now works correctly)
[WLANCONF]
- Use the correct length for NDIS requests
- Indicate success to the console for -c or -d
- Print current association info if called with no parameters
svn path=/branches/wlan-bringup/; revision=54897
- Apply the same hack for OID_GEN_MEDIA_CONNECT_STATUS as we did for OID_GEN_HARDWARE_STATUS (since media state detection is broken for certain adapters)
[DHCPCSVC]
- Keep the adapter thread running to handle media state changes
[TCPIP]
- Fix media state detection which was totally broken because it was using the wrong OID
svn path=/branches/wlan-bringup/; revision=54895
- Use a different method for determining if an adapter supports 802.11 networks so it works on drivers that don't implement OID_GEN_PHYSICAL_MEDIUM
svn path=/branches/wlan-bringup/; revision=54888
NTOSKRNL/LPC] Don't check if ReplyMessage is NULL inside SEH, since it's dereferenced later anyway
[WIN32K] Don't try to print uninitialized variables
Patch by Samuel Serapion
svn path=/trunk/; revision=54881
- Use the correct IOCTL input buffer
[WLANCONF]
- Fix parameters to IOCTL_NDISUIO_QUERY_BINDING
- Wlanconf is ready for testing with a real WLAN adapter (for anyone who wants to checkout this branch and try it)
- Run "wlanconf -s <SSID>" to connect to an unencrypted wireless network
- Run "wlanconf -s <SSID> -w <WEP key>" to connect to a WEP encrypted wireless network (WPA not supported)
svn path=/branches/wlan-bringup/; revision=54875
- Add some definitions
[WLANCONF]
- Add support for adhoc networks and WEP encryption
- Fix the BSSID info loop
svn path=/branches/wlan-bringup/; revision=54874
- Initial drop of a little WLAN configuration tool to use until WZC is implemented and working
- Doesn't build yet due to missing OID definitions
- Some of this code will eventually be reused when implementing WZC
svn path=/branches/wlan-bringup/; revision=54869
- Add an entry for ndisuio (temporary until INF install works)
- Autostart WZC service
[NDISUIO]
- Fix crash during binding and receive
- Add more debugging
svn path=/branches/wlan-bringup/; revision=54857
- Fix few leaks and possible buffer overflows in New menu code
- Add icons of folder and shortcut to New menu
- Display message if file can't be created
svn path=/trunk/; revision=54848
- Rewrite Open With menu and dialog
- Load all applications
- Use treeview with Recommended and Other programs
- Don't allow duplicates
- Load registry keys in proper way
svn path=/trunk/; revision=54844
- Implement quering compatible ids
- Fix result status for quering instance id
- Fix bug in msmouse.inf
svn path=/branches/usb-bringup/; revision=54843
- Add new style SAL definitions (only dummys for now)
- Move old style SAL definitions to sal_old.h
- Convert annotations in sspi.h to new style
- Allow new style annotations for C++ again, old style annotations are not possible with C++ and gcc headers
svn path=/trunk/; revision=54839
- Add a ReactOS hack to let the hid stack initialize
- Add input.inf
- HID Mouse initializes and starts, now win32k needs some work to open the mice
svn path=/branches/usb-bringup/; revision=54838
- Fix creating instance id
- Print debug message
- Sync msmouse with head and add hid installation
- Needs input.inf written
- HID stack starts to initializes and then fails the to enable the device interface (/drivers/hid/hidclass/pdo.c:511), Logs show that IoGetDeviceObjectPointer() fails with STATUS_UNSUCCESSFUL
svn path=/branches/usb-bringup/; revision=54837
- Create a pesudo random instance id
- Fixes assert when plugin usb devices whose device descriptor don't provide a serial number
svn path=/branches/usb-bringup/; revision=54836
- Implement a HID parsing library, which is based on Haiku`s exellent HID Stack
- Implement HidP_FreeCollection, HidP_GetCaps, HidP_GetCollectionDescription, HidP_MaxUsageListLength, HidP_GetSpecificValueCaps, HidP_GetUsages, HidP_GetScaledUsageValue
- hidparse driver is now implemented enough to support a hid mouse
- Tested in VBox 4.1.4 + WinXP + ReactOS usbohci+ mouhid+ hidclass+ hidparse+hidusb
svn path=/branches/usb-bringup/; revision=54834
Complete RtlpDosPathNameToRelativeNtPathName_Ustr() rewrite by properly handling relative names.
This fixes several regressions brought by r54804 (see winetests).
This also fixes file system corruption when using directory functions.
This may also fix general file system corruption, so, please retest bug #6720.
svn path=/trunk/; revision=54820
- Handle IOCTL_NDISUIO_OPEN_WRITE_DEVICE and IOCTL_NDISUIO_OPEN_DEVICE properly with regard to current open handles
- Check that we have read permissions (opened by IOCTL_NDISUIO_OPEN_DEVICE) before allowing queries or reads
svn path=/branches/wlan-bringup/; revision=54819
- Fix a race condition during adapter context destruction
- Add missing IRP completion in IRP_MJ_READ and IRP_MJ_WRITE handlers
- Create an adapter context in response to PnP events
svn path=/branches/wlan-bringup/; revision=54816
- add module module type, which is basically a DLL without entry point and use it for typelibs
- get rid of custom targets for generating .tlb files and use source file dependencies instead
svn path=/trunk/; revision=54813
Comment out two series of tests untill a proper fix is available. Those lead to memory misuse in shell32.
This has been revealed by r54804.
Should fix both test bots.
svn path=/trunk/; revision=54811
- Add the NDISUIO driver to allow communication between user-mode and kernel-mode NIC drivers
- Only the IOCTLs required for WLAN are implemented right now
- Initial source drop without build testing
svn path=/branches/wlan-bringup/; revision=54810
- Create PDO for each top level collection found
- Use the specified collection index instead of the first
- add collection id to hardware id when the device has more than one top level collection, see http://msdn.microsoft.com/en-us/windows/hardware/gg487473 for details
svn path=/branches/usb-bringup/; revision=54808
- Fake status success for unimplemented IRP_MN_REMOVE_DEVICE
- Fix tons of bugs in the read report completion routine
- Implement function to re-use a complete report irp
- Use correct device object when invoking the mini driver's dispatch routine in the read function
- silence a few traces
- Mouse now works with with ReactOS USB stack + ReactOS HID stack (TBD: implement hidparse.sys)
svn path=/branches/usb-bringup/; revision=54806
Reimplement CreateDirectoryA, CreateDirectoryExA, CreateDirectoryW, CreateDirectoryExW, RemoveDirectoryA, RemoveDirectoryW.
This reimplementation now matches the w2k3 one, properly handles reparse points (especially mount points) and also brings several fixes.
svn path=/trunk/; revision=54804
- Partly implement IRP_MN_START_DEVICE for PDO
- PDO now completely initializes
- Remaining IOCTL to implement IRP_MJ_CREATE, IRP_MJ_CLOSE, IRP_MJ_FLUSH, IRP_MJ_READ for partial mouse support in hidclass
svn path=/branches/usb-bringup/; revision=54792
CreateDirectory: handle the case when the FS doesn't support EAs.
This should fix VBox and is mandatory if we want to support MS FSD with FAT32
svn path=/trunk/; revision=54789
- Remove an unneeded (and prone to race) RtlZeroMemory (found by Thomas Faber)
- Fix misuse of PAGE_MASK macro and broken PTE flag comparisons
svn path=/trunk/; revision=54782
- Implement AddDevice routine for hidusb
- Fix bug in Hid_DispatchUrb, which returned not initialized status code
- Implement IOCTL_HID_GET_REPORT_DESCRIPTOR, IOCTL_HID_READ_REPORT
- Add support routines for reseting / aborting pipe, not yet used
- TBD: implement error handling
- hidusb now works in combination with a usb mouse (other hid devices such as keyboards will require more ioctl implemented)
- Tested in Vbox 4.1.4 + WinXP + ReactOS usbhid.sys + ReactOS usbmou.sys
- Testing procedure -> inject usb mouse after startup (needs mouse integration disabled -> thanks to timo for pointing this out)
svn path=/branches/usb-bringup/; revision=54781
- Silence a few traces
- Add a hack for handling URB_FUNCTION_CLASS_INTERFACE. The hid bus driver performs this request with zero buffer length, which is not valid. Needs more investigation
svn path=/branches/usb-bringup/; revision=54777
- Usage page of zero is not used, increment, the MouHid_ButtonUpFlags / MouHid_ButtonDownFlags array
- Mouse clicks now work
- Implement flushing of the hid report queue
- Driver now works in VBox 4.1.4 + WinXP
- Secondary mouse pointer does not move in VBox (either a Vbox bug / WinXP), tested with ms driver and the same results were reveiled
svn path=/branches/usb-bringup/; revision=54776
- Fix up / down button detection
- Fix move detection. Mouse changes are now detected
- Mouse does not yet work in Vbox, as mouclass driver is for unknown reasons not connecting to the mouhid, needs to be investigated
- Tested in VBox 4.1.4 + ReactOS mouhid.sys
svn path=/branches/usb-bringup/; revision=54775
- Fix a few bugs & race condition in the read report routine
- mouhid initializes and is able to read input reports
- button press / wheel state change is detected
- mouse move detection not yet working
svn path=/branches/usb-bringup/; revision=54774
- Tweak the balancer to prevent an excessive buildup of user pages while cache is paged out completely each run
- Bugcheck if we've trimmed everything as much as possibly but we are still in need of pages
svn path=/trunk/; revision=54771
- Implement mouse button change detection
- Implement support routine for dispatching mouse input data
- Allocate mdl for input report, not yet used
svn path=/branches/usb-bringup/; revision=54767
- Don't flush the TLB in MmEnableVirtualMapping and MmCreateVirtualMapping because the PTE is already invalid so it can't possibly be in the TLB per x86 spec
- Added explanations of why flushes are needed or not needed in certain places
- A few small code improvements
- Should speed up ROS a bit by avoiding unnecessary TLB entry flushing
svn path=/trunk/; revision=54766
- Replace the log file critical section by a resource and use shared locks for read actions and exclusive locks for write actions.
- Add stubs for clear and backup actions.
svn path=/trunk/; revision=54763
- Reference the cache segment when flushing to avoid a race between deleting the file cache and the balancer flushing pages to disk which causes random cache segments to be freed while we're holding a lock on it
- Fixes a crash in KeReleaseMutex (used to be a crash in ExReleasePushLock) during periods of high memory load
svn path=/trunk/; revision=54745
- Samuel Serapion: Implement RtlIpv4AddressToStringEx A/W and minor enhancements RtlIpv4AddressToStringA/W. Fixes some ntdll:rtl winetests.
See issue #6779 for more details.
svn path=/trunk/; revision=54742
- Wait to acquire locks in CcRosFlushDirtyPages
- This is needed because this path is used during critical times like shutdown and very low memory situations when flushes cannot fail
svn path=/trunk/; revision=54725
- Remove "Explore from here" from context menu. It's useless, does not exist in Windows and can't be translated in current form.
svn path=/trunk/; revision=54723
- Fix broken code in drive.cpp. Spotted by r3ddr4g0n.
- General code improvements
- Don't try to create static open handler in registry when file properties are opened. Fixes two "Open" items in shortcuts context menu after opening properties window.
See issue #6746 for more details.
svn path=/trunk/; revision=54722
- Fix SHCreateSessionKey
[EXPLORER]
- Don't run startup applications if StartupHasBeenRun key exists. If it doesn't, create new key.
- Make startup code readable
See issue #1801 for more details.
svn path=/trunk/; revision=54715
- A number of fixes to CmpQueryKeyName():
* Properly check the provided buffer size against needed size.
* Don't overwrite user provided buffer.
* Write as much data as could fit into the buffer (this is normal behaviour for any query function in the kernel), returning STATUS_INFO_LENGTH_MISMATCH if not all data were written.
Thanks to r3ddr4g0n for identifying the problem, testing with DPH and testing this patch.
svn path=/trunk/; revision=54711
- Properly close Open With dialog window (EndDialog works only for modal dialogs, modeless dialogs should be destroyed with DestroyWindow instead). Fixes folder window hang if Open With windows was opened twice (and also a window leak).
svn path=/trunk/; revision=54701
- "Hyperspace is local!": OK, system PTE space isn't…
- Change the function name to reflect the real location of the mapping and match its partner MiUnmapPagesInZeroSpace()
svn path=/trunk/; revision=54700
- Lock the registry before flushing hives to avoid, among other things, use of an ERESOURCE without being inside a critical region
svn path=/trunk/; revision=54694
- Tweak the balancer a bit to favor RosMm allocations in low memory scenarios
- Fix the balancer out of memory warning
svn path=/trunk/; revision=54693
- Only unmap pages if they were mapped
- Make sure the page reference counts are sane when freeing the cache memory area
svn path=/trunk/; revision=54690
- Fix getting name for context menu item. Fixes two "find" items in My Computer context menu.
- Minor code and comments improvements
svn path=/trunk/; revision=54688
- Fail if display name device is given to EnumerateDisplayDevices and index is not set to 0. Fixes hang in Display Settings -> Advanced -> Monitor.
svn path=/trunk/; revision=54679
- Revert 54075 hack (registration from old shell32 based on WINE), so things does not get registered twice. Two Control Panels in MyComputer are no longer visible
- Fix all regressions, which was hack-fixed before.
- Add more debug output in case of errors
svn path=/trunk/; revision=54673
- Implement ElfrReadELA.
- Improve LogfReadEvent in such a way that it can return ANSI and UNICODE log entries.
- Add a hack to ElfrReadELW in order to return a proper status code in case of an end-of-file situation.
svn path=/trunk/; revision=54672
EngAllocMem:
- Respect the FL_NONPAGED_MEMORY flag (actually allocate the memory from non-paged pool)
- Fix an improper flag comparison that caused memory allocated with both FL_NONPAGED_MEMORY and FL_ZERO_MEMORY set to not be zeroed as requested
svn path=/trunk/; revision=54669
- Sync translated resource files (DeskADP and most of CPL) with English files.
- Small Czech update of DXDiag and TaskMgr translations by Radek Liska (bug #6774).
svn path=/trunk/; revision=54663
Follow original design/style of printf module
Always use dosmaperr to convert win32 errors to sane crt errors
Enable forgotten _get_doserrno, get_errno, set_doserrno and _gmtime32_s
Small fixes to pass more wine tests.
Patch by Samuel Serapión.
svn path=/trunk/; revision=54661
- Fix SxS path calculation, now it returns proper full path to the DLL instead of just "C:\ReactOS\winsxs". Spotted by AbiWord 2.9.2.
svn path=/trunk/; revision=54659
- inetcpl romanian translation update, resource file update for all languages by Stefan Fulea (bug no. 6766);
- polish translation update of inetcpl and resources changed by revision 54644;
svn path=/trunk/; revision=54657
- Implement ElfrClearELFA and ElfrClearELFW.
- Fix the module name conversion in ElfrOpenELA and ElfrRegisterEventSourceA.
svn path=/trunk/; revision=54656
Nothing calls this code yet (SearchPathW soon will), this is just an implementation commit. Code was tested to boot to 3rd stage after SearchPath modification, which is the next target (step 2).
After which, the only code in the loader paths that uses legacy path APIs will be the create process path, the third and final step.
svn path=/trunk/; revision=54646
- Handle SO_SNDBUF in WSPSetSockOpt so that it doesn't fall through to wshtcpip and result in "unknown level: 65536".
- Added more trace messages.
See issue #5477 for more details.
svn path=/trunk/; revision=54643
ReportEventA/W: Strings must be passed to ElfrReportEventA/W as a pointer to an array of PANSI_STRINGs instead of a pointer to an array of ANSI_STRINGs. Now, strings appear in the event viewer.
svn path=/trunk/; revision=54642
- Fix Get/SetWindowPlacement and everything related. Which exposed the scroll bar over write issue, it's not refreshing the client window when it is maximized. Next thing to fix.
- Fix ShowWindowAsync, now restore and minimize all works. This is based on wine but in reality it's based on Windows queuing event messages used in server side internal calls. Also this patch includes the server side hooking points.
- Removed one ugly API.
- Fixes bug 6239, 6739 and client window restore from maximize and the incremental crawling down issue when parent window is minimized then restored.
svn path=/trunk/; revision=54637
- Fixups to set focus and show window support, added notes. Add is window visible. Fix a binary bit return, 1 is TRUE. Fix get client rectangle. Minor fixups.
svn path=/trunk/; revision=54634
- Cache the last data block and offset so we don't have to search from the start each time
- This avoids needless swapping of cabinet data in and out of RAM
- File copy in text-mode setup on 32 MB RAM is down from 8 minutes to 20 seconds (in my testing)
svn path=/trunk/; revision=54632
- Fix invalid errors about "system DLL kernel32.dll was relocated in memory". This reverts my previous fix suggested by rafalh and puts a better explanation there in comments.
svn path=/trunk/; revision=54629
- Improve and rewrite a few active context support functions in RTL, making them more close to their real functionality and throwing away wineisms.
- Uncomment a line in RtlAllocateActivationContextStack() which was causing it to always allocate a new activation context stack even if there was one. Thanks to my previous commit in kernel32 counterpart, this hack is not needed anymore and memory won't be wasted.
- Since we have mostly Wine's code in our activation context implementation, it makes sense to use it properly until it's rewritten, so first of all:
- Properly initialize current process's activation context support by calling actctx_init() from LdrpInitializeProcess instead of calling it all the time when a new context is created.
- Bring in create_module_activation_context and find_actctx_dll (newer version from Wine which supports manifests/assmeblies in local directories) and use it. Every place where it's used is marked as a HACK so it will be easy to detach it later.
- Remove actctx-related hack LdrpInitializeThread, now it works properly.
- Fix a typo in LdrpMapDll found by rafalh. Now the error-detection condition will be correctly evaluated.
See issue #6611 for more details.
svn path=/trunk/; revision=54623
- Zero-initialize ActivationContextStack pointer before calling RtlAllocateActivationContextStack(). This allows to get rid of some hacks in the existing ActCtx support in the ntdll loader and actual implementation.
svn path=/trunk/; revision=54618
- Don't strip relocations for some sections. Linker knows better if they are needed. Fixes removed relocations for delayed imports in idata section. Bug traced by me, Timo and Amine.
See issue #6618 for more details.
svn path=/trunk/; revision=54613
- Massive IPPacket usage overhaul (fix missing frees, initializations, and double frees)
- Make IP receive completely agnostic of the buffers and data location (part 2 of 2)
- Use paged pool for the majority of allocations during receive and assembly
- Implement ProtocolReceivePacket for faster performance with 3rd party NIC drivers
- Fix a buffer overflow if an IPv4 packet is received with a header larger than 20 bytes
- Many more improvements and bug fixes
svn path=/trunk/; revision=54597
- Fix more set active window API tests. The window postion change message with fixed lparam is erroneous. How can the address be the same since it originates from kernel space?
svn path=/trunk/; revision=54596
- Avoid an extra copy operation when receiving packets
- Optimize loopback code a bit
- Make IP receive independent of the location of the data and continuity of buffers in the NDIS packet for ProtocolReceivePacket support (part 1 of x)
svn path=/trunk/; revision=54593
[KERNEL32]: Replace BasepGetDllPath with BasepComputeProcessDllPath. Paths now work right. This is step 1. SearchPatchW is next (BasepComputeProcessPath), followed by CreateProcessInternalW (BasepComputeProcessExePath). Then all paths will work right.
svn path=/trunk/; revision=54592
- Fix a logic error
- Signal the balancer in the !CanWait case if we're too low on pages since it runs asynchronously as of recently
svn path=/trunk/; revision=54591
- Prevent buffer overflow when creating the display value for a value of type REG_MULTI_SZ.
- Prevent processing excess data when preparing a value of type REG_MULTI_SZ for editing.
svn path=/trunk/; revision=54590
- Print a warning and signal RosMm to trim pages if we drop below MmMinimumFreePages in MiRemovePageByColor and MiUnlinkFreeOrZeroedPage
svn path=/trunk/; revision=54589
- Depending on how big the full path is, either the application path or the full path should be used. Size the buffers correctly, too.
- If the caller did not specify a window title, should use the app path as title.
- Clear the current directory handle if InheritHandles is false.
- Correctly handle CREATE_NEW_PROCESS_GROUP and CREATE_NEW_CONSOLE.
- Check the failure codes of many APIs that we just assumed returned success.
- STARTF_USEHOTKEY | STARTF_SHELLPRIVATE should be checked when inheriting console handles.
- Handle 1MB and 16MB reservation requests.
- Bugs remain: ImageSubsystem not written, and failure will leak. To be fixed later.
svn path=/trunk/; revision=54587
- Convert the remainder of PL strings to UTF-8 to fix build, since that RC is declared as Unicode. Convert italian resources to UTF-8 and place all rc in a single section;
svn path=/trunk/; revision=54582
- Release the address space lock before entering SEH because an exception would result in recursive acquisition of the address space lock when the fault handler tries to grab it
svn path=/trunk/; revision=54579
Remove class2 and scsiport service keys from the registry because these drivers are loaded automatically when they are needed.
svn path=/trunk/; revision=54575
- store property item when there are custom specific properties
- Fix bug where the property handle routine received an invalid pointer to the data
svn path=/trunk/; revision=54572
add wine dll registration functions to libwine (taken from winecrt0)
TODO: support for linking multiple res files
rbuild/cmake: generating .rgs files from typelibs
svn path=/trunk/; revision=54566
NtFreeVirtualMemory:
- Handle the case where a region size of 0 is passed in
- Return the correct error status for failure
- Copy back the rounded values
- Add checks and prints to catch callers doing nasty things (one is commented out because csrsrv triggers it trying to release the first 1 MB of RAM during video init, not sure what to do there)
- There is a heap bug where calling RtlReAllocateHeap on a large block allocation (HEAP_ENTRY_VIRTUAL_ALLOC is set) will cause a call to NtFreeVirtualMemory with an offset base (illegal) and a length smaller than the total pages reserved in the VM region (also illegal). This bug is exposed by setupapi when it parses large INF files (like the PRO/1000 driver INF).
svn path=/trunk/; revision=54565
Normalize all crt dll entry points to be identical.
Implement a single per thread data structure, instead of ancient and incomplete
"wine port" hack and use it everywhere.
Remove completely unused HeapCreate calls that just wasted memory.
Update cpp.c and cppexcept.c to wine 1.3.33
Fix up strtok, wcstok, rand, and _beginthread_trampoline according to wine.
Add place holders for future locale stuff.
Patch by Samuel Serapion.
svn path=/trunk/; revision=54563
- Implement NdisReturnPackets and do proper reference tracking of packets sent to ProtocolReceivePacket
- Plug a massive memory leak that resulted in leaking every packet descriptor that any deserialized miniport indicated to TCP/IP
svn path=/trunk/; revision=54559
- Call FreeCommonBuffer() synchronously if we're running at PASSIVE_LEVEL to avoid cases where the miniport could be freed before the work item runs
svn path=/trunk/; revision=54558
- Fix regressions, running RegEdit. Adding more support for scroll bar controls. Pass all but 7 tests including the two wine todos.
svn path=/trunk/; revision=54556
- Fix scrollbar class support. Sync ports from wine. Added the window scroll structure to the class. Pass all but 15 tests, two are wine todos lines 304 and 343.
svn path=/trunk/; revision=54555
- Optimize CcRosTrimCache by eliminating an extra loop of all segments
- Allow swapping of dirty segments by flushing them first
svn path=/trunk/; revision=54549
- Add NtAllocateVirtualMemory test, which is a small stress test for virtual memory allocation/freeing/reuse
- Can be used to reproduce bug 5857 in third stage (install with VT-x enabled, then disable for the test)
- Shows that the bug is a race condition in Mm, and not heap's fault
- I've put ASSERTs instead of ok's for easier debugging
svn path=/trunk/; revision=54548
- Many balancer fixes for concurrency, improved swapping efficiency, and reduced code duplication
- Move the low memory case back under the PFN lock
- Debugging is on for now because I don't trust the paging code all that much (it was not used very much until recently)
svn path=/trunk/; revision=54544
- Revert part of r54530
- Try to rebalance RosMm's pages if we're low on memory in ARM3
- ROS runs with 48 MB RAM again (but setup is very slow due to horrendous paging caused by cache pollution)
svn path=/trunk/; revision=54539
- Don't signal the balancer from within the balancer
- Optimize usage of the balancer a bit so we don't have to wait on it as much
svn path=/trunk/; revision=54537
MSPaint:
- Add Swedish translation by Michael Wu (bug #6695).
- Sync all translation files with American English file (bug #6700).
svn path=/trunk/; revision=54533
- Display file size in bytes in file properties dialog
- Don't use ULONGLONG if ULONG is enough
- Change LARGE_INTEGER to ULARGE_INTEGER where it makes sense
- Fix calculation of file size in CountFolderAndFiles
- WM_GETTEXT receives size in bytes (not characters count)
- Use Get(Set)DlgItemTextW instead of SendDlgItemMessageW
Patch by Carlo Bramix. Improvements by me.
See issue #6294 for more details.
svn path=/trunk/; revision=54529
All .lnk files in Windows has structure called "id list". This is something like a path (we can't use environment variables there) and it is relative to Desktop. Usually first element is My Computer folder. In our shortcuts it points to livecd_start.cmd in My Documents folder, which is simple batch file. It's functionality is to start application given in argument. Because command line parameter of shell links can have environment variables, our shortcuts does not depend on patch to system directory (especially the drive number). It's a bit complicated but I have not found a better way.
Now we have working shortcuts on Desktop and Start Menu. Control Panel will benefit in future because Admin Tools depend on Start Menu.
svn path=/trunk/; revision=54514
Switch to the 'new' disk geometry (63 sectors per track and 255 tracks per cylinder). This geometry is compatible with Windows, Linux and other Operating Systems.
ATTENTION: You must delete and re-create the partitions on your harddisk.
svn path=/trunk/; revision=54511
- Enable translations of the language page. Patch by Stefan Fulea.
- Simplified the clear text code.
- Fixed the german language page.
See issue #6637 for more details.
svn path=/trunk/; revision=54502
- Fix errno in dialog.c by linking to msvcrt with rbuild. Cmake does not have this issue, already links to msvcrt. Gdi32 rbuild does link to crtdll instead of msvcrt. Should we fix this too, before the Cmake switch over to see anything is different?
svn path=/trunk/; revision=54498
- Fix a typo in NtUserSetClipboardData that tried to print a struct as a pointer (cid 15051)
- Clean up clipboard.c a bit. note: this cleanup "fixes" cid 14599 and 14598, but even though they are false positives, this cleanup was needed
- Add missing break in NtUserQueryWindow (cid 13486)
svn path=/trunk/; revision=54483
- Fix dwServiceType and dwServiceState checks in REnumServicesStatusW (see r54373)
- Fix a memory leak in RQueryServiceConfig2A
svn path=/trunk/; revision=54481
- Also round down offset we're given when creating a page op, so we don't miss the one that could be running at the time
svn path=/trunk/; revision=54480
- create a page op when changing page protection attribute
- properly handle PAGE_NOACCESS protection change in NtProtectVirtualMemory
- Round down the address we're given when creating a page op, so we're sure there is o,ly one page op for running for one page
- Add a small hack
Now ReatOS boots with DPH enabled on low memory conditions
svn path=/trunk/; revision=54477
- Fix process reference leak in case of GuiConsoleOpenUserRegistryPathPerProcessId failing to open registry key. Fixes console processes in LiveCD being ghosts.
svn path=/trunk/; revision=54475
- Do not repeat messages into serial port even when using other debug methods.
- Allow to specify debug log file name.
- When using the "Screen" method, show blue screen when entering KDB.
- Disable the repetition of previous KDB command with long many-page output.
[USETUP]
- Add "ReactOS (Screen)" entry to boot menu.
svn path=/trunk/; revision=54468
- Print debugging information if a mismatch occurs between requested and returned parameter types
- Print debugging information if a registry read returns a type that NDIS doesn't handle
- Ensure that all digits are less than the base during conversion from string to integer (otherwise it's a string)
svn path=/trunk/; revision=54464
Break ReactOS CRT:
- Add back crash on NULL pointer arg. This is the behaviour on w2k3.
- Don't skip \\?\ prefix any longer. This is the behaviour on w2k3.
svn path=/trunk/; revision=54452
- Display all hardware profiles
- Add the hardware profile copy dialog to most languages.
- Disable the Up, Down, Property, Copy, Rename and Delete buttons in the Hardware Profiles dialog.
svn path=/trunk/; revision=54448
- Bring more awesomeness to the heap manager by zefklop's request:
* Add support for settable user values and flags in RtlAllocateHeap and RtlReAllocateHeap.
* Return error if RtlSetUserValueHeap was unable to set the value.
* Fixes remaining 2 failures of the kernel32_winetest alloc test.
svn path=/trunk/; revision=54434
Fix _fpclass prototype
Fix headers included
Remove internal defines that were useless and even not matching real defines.
This fixes 4 tests
svn path=/trunk/; revision=54432
- Updating MsgiAnsiToUnicodeReply fixed bug 4856 and 6650. Mirrored with it's counterpart. More work is needed.
- Adding some future DBCS char support.
- Miscellaneous changes.
svn path=/trunk/; revision=54428
Add test suite for IEEE floatting-point functions.
It's been tested on w2k3 and it has 0 failed tests.
Some are failing on ReactOS.
svn path=/trunk/; revision=54425
- Implement support for hot plugging PS/2 mice if one was present at boot (same requirement as Windows)
- Fixes bug 1395
svn path=/trunk/; revision=54415
- Make sure GetEnvironmentVariableW does not use uninitialized variable
- Fix GetDllLoadPath not terminating string with NULL if PATH env variable is not defined. Fixes hang on manual INF selection in New Device wizard.
See issue #6480 for more details.
svn path=/trunk/; revision=54406
- Fix a critical bug in registry entry parsing that was causing all REG_SZ values to be reported as NdisParameterTypeInteger
- Distinguish between NdisParameterTypeInteger and NdisParameterTypeHexInteger
- 3Com NICs are fully functional now [bug 4330]
svn path=/trunk/; revision=54402
- Fix running the test with a relative path. Already Sent Upstream(TM)
- Do the rest of that tiny Winesync, too, while we're at it. Now at 1.3.33
- Remove the spooler hack in the process. We don't actually need it
svn path=/trunk/; revision=54400
- Romanian translation update (bug no. 6654) by Stefan Fulea;
- Simplified Chineese translation of base/shell and system (bug no. 6655) by Song Fuchang;
- Slovak translation fixes (bug no. 6658) by Mario Kacmar;
svn path=/trunk/; revision=54399
- Some more cleanup.
- Work with page rounded down addresses in SectionView page faults.
- You should not create a virtual mapping on a non page-aligned address range : ASSERT about that.
- Flush Tlb each time we change it
- Fix logic I broke in MmCreateVirtualMapping (thanks to Thomas Faber)
svn path=/trunk/; revision=54392
- ensure that all callers of MM providing an user land address also provide a process.
- ASSERT about that
- Fix broken check
svn path=/trunk/; revision=54390
- call mm functions with a process, when we have one.
- Fix potential rounding issues
- Add some sanity ASSERTs
- You never use enough brackets ;-)
svn path=/trunk/; revision=54386
- Remove broken map register "prefetcher" code
- Check that we have enough space to grow our map registers
- Fix a bad hint to RtlFindClearBitsAndSet
- Queuing the grow work item is non-vital; treat it as such
- Fixes BSOD with CM3738 and VIA1723 sound cards [bug 6657]
svn path=/trunk/; revision=54382
- Always return ERROR_SUCCESS when ScmWaitForServiceConnect completes successfully instead of sometimes returning ERROR_PIPE_CONNECTED because the callers of this helper just look for ERROR_SUCCESS to indicate a successful call
svn path=/trunk/; revision=54381
- Export IContextMenu interface in RecycleBin object. Fixes recycle bin context menu.
- Export IShellPropSheetExt interface too and fix order.
See issue #6543 for more details.
svn path=/trunk/; revision=54379
- Fix a critical bug in NdisMGetDmaAlignment which was causing every NIC driver calling this function to crash because we thought we were passed an NDIS_DMA_BLOCK handle when in fact we receive an NDIS_MINIPORT_BLOCK handle
- Fixes RTL8169 and 3Com NICs (and probably several more) [bug 6651 and bug 4330]
svn path=/trunk/; revision=54372
- Fix DMA on the EISA bus
- Fix a critical calculation error in map register size resulting in the map register allocation worker not allocating enough map registers for a dynamic resize
svn path=/trunk/; revision=54371
- base/setup,shell and system conversion patch, by UTF-8 conversion patch by Elton Chung, see bug 6333;
- Romanian translation fixes and UTF-8 conversion by Stefan Fulea, see bug 6603;
svn path=/trunk/; revision=54364
[KERNEL32]: Add notes on how DeviceIoControl and Beep should support TS scenarios in the future (nod to Ged).
[KERNEL32]: Make Beep call NotifySoundSentry, and implement the latter, which sends a message to CSRSS w.r.t the previous commit. Obviously we are not anywhere close real sound sentry support, but at least part of the path is there now.
svn path=/trunk/; revision=54358
- Allow changing the image path with ChangeServiceConfig. Fixes timeout that shouldn't actually be a timeout in advapi32:service test.
See issue #6646 for more details.
svn path=/trunk/; revision=54348
- Disable usbdrv
- I've heard too many reports of it breaking boot and too few reports of it actually working on real hardware (although it does work perfectly in QEMU)
svn path=/trunk/; revision=54347
- Skip service start timeout tests. They cannot time out until services uses asynchronous I/O.
See issue #6646 for more details.
svn path=/trunk/; revision=54346
- Do not use local memory functions (and possibly free the result using HeapFree!)
- Do not use double percent signs with _wcsnicmp
- DPRINT and style fixes
svn path=/trunk/; revision=54344
[KERNEL32]: Rewrite the tape APIs to use a single generic function, and in the process support pending I/O now, which wasn't handled before.
[KERNEL32]: Moved the interlocked APIs to synch.c instead of dllmain.c
svn path=/trunk/; revision=54334
- Fix a potential CCB state inconsistency if the IRP is canceled before adding it to the pending list
- Don't wait if the IRP has been canceled
- Preserve the return status of NpfsAddListeningServerInstance
svn path=/trunk/; revision=54328
[KERNEL32]: Cleanup files a bit.
- Bintype.c is gone and moved into VDM (these are functions for VDM support).
- Npipe.c and pipe.c are merged.
- Dosdev is moved into client, these are not quite File APIs.
- Base(8Bit/Unicode) APIs are moved into utils.c
- Wow64 APIs are moved into utils.c
- The mess that was file.c has the Replace* file APIs going into move.c (similar behaviors), the Open* APIs into create.c (similar) and the Read/Write Scatter/Gather APIs into rw.c (where they belong). What remained in file.c were mostly Set/Query functions, so it becomes fileinfo.c
- The remaining APIs in file.c were all about paths -- at the file level, specifically filenames, so they go into filename.c
- Move some File API functions from certain files into other files where they belong more.
svn path=/trunk/; revision=54326
- Make the initial NpfsRead wait uninterruptible since it's really being used like a mutex (avoid the kernel stack being paged out during the wait causing memory corruption)
- Handle interrupted waits that were missed in certain cases
- Fix a potential deadlock that would happen if more I/O was done on a file object after the previous operation was interrupted by an APC or alert
svn path=/trunk/; revision=54325
[KERNEL32]: Reimplement GetFullPathNameA to implement similar functionality and checks as the other reimplemented A functions, especially w.r.t ANSI/OEM and DBCS strings.
svn path=/trunk/; revision=54321
- Fix wait parameters so we can break out of a wait if a user APC is delivered and the file object is opened with FILE_SYNCHRONOUS_IO_ALERT
- Fixes hanging ntdll:pipe winetest (only 1 failure out of 53 tests now)
svn path=/trunk/; revision=54319
- Windows allows the two input buffers to overlap. This means we have to be very careful with copying (and always use RtlMoveMemory) -- the old function was not handling this at all.
- We also have to handle cases where we need to make our own local buffer copy.
- Length validation is more stringent now.
- Checking for short/long path names wasn't correct w.r.t ANSI file mode, as it was calling Rtl which assumes OEM.
- Shortcuts were taken while parsing slashes and path separators. We now call into Rtl to support this, and also support unlimited slashes (note that \\??\c:\\\windows\\\\system32 is actully a valid path, for example).
- ErrorMode is now correctly set for the thread, to avoid "Insert floppy" or "Close CDROM bay door" errors as we are using the FindFile API.
- Correct LastError is set where appropriate.
- An application compatibility flag is now supported.
svn path=/trunk/; revision=54318
- Use the correct buffer length in GetDllDirectoryA
- Do not always return FALSE from BasepIsCurDirAllowedForPlainExeNames
- Some cosmetic fixes
svn path=/trunk/; revision=54317
- Use FIELD_OFFSET for variable-length structure sizes
- Handle an invalid parameter condition in RCreateServiceW. Fixes an advapi32:service test
- Do not dereference a NULL-pointer on out-of-memory
svn path=/trunk/; revision=54315
[KERNEL32]: Fix SetDllDirectoryW:
- Fail if ";" is in path
- Use RtlCreateUnicodeString instead of manual string management -- if the allocation fails, do not free the current base directory string! Older API would always free the current directory string, even if setting up the new one failed.
- Don't assume lpPathName is filled out, it can be NULL, and this means the DLL directory should be cleared.
[KERNEL32]: Fix SetDllDirectoryA as above, additionally use RtlInitAnsiStringEx to protect against overflow, and use Basep8BitStringToUnicodeString, to add support for OEM paths instead of assuming ANSI.
[KERNEL32]: GetDllDirectoryW: correctly NULL-terminate and check lengths
[KERNEL32]: GetDllDirectoryA: same as above, plus use BasepUnicodeString* APIs to correctly support OEM paths instead of assuming ANSI.
[KERNEL32]: NeedCurrentDirectoryForExePath(A/W): call internal BasepIsCurDirAllowedForPlainExeNames instead of converting from A->W, this isn't needed.
[KERNEL32]: GetFullPathNameW just calls RtlGetFullPathName_U, no extra checks/etc are needed.
Next steps are GetFullPathNameA, GetShort/LongPathNameA, GetShort/LongPathNameW, hoping to bring the wine test failures closer to zero.
svn path=/trunk/; revision=54309
- Initialize OLEINPLACEFRAMEINFO.cb for IOleInPlaceSite::GetWindowContext. Fixes explorer "Web" button assert. Already applied upstream.
See issue #6011 for more details.
svn path=/trunk/; revision=54304
- Fix ParseDisplayName based on wine by checking for paths that are parsable through IEParseDisplayNameWithBCW. Score 2 failed wine tests in shlfolder
svn path=/trunk/; revision=54300
- Sync heap.c and undname.c with Wine 1.3.32
- Remove incorrect casts. Undo an incorrect warning fix. Fixes hang in msvcrt:cpp test
svn path=/trunk/; revision=54297
- Fixed Dialog Ansi and color control brush acquisition.
- Added stack protection for message recursion. Due to wine message bug when in dialog initialization. See bug 6126.
- Miscellaneous fixups.
svn path=/trunk/; revision=54293
[KERNEL32]: Acquire SE_INCREASE_WORKING_SET_PRIVILEGE in SetInformationJobObject if needed.
[KERNEL32]: Only do conversion of Process Priority Class in SetInformationJobObject if the limit flags request it.
[KERNEL32]: Add missing size checks in QueryInformationJobObject.
[KERNEL32]: Use RtlEncode/DecodePointer to protect the top level exception handler pointer, as done on NT 5.2+
[KERNEL32]: Fix IsBadRead/WritePtr to use the correct page size from the CSRSS shared section.
[KERNEL32]: Fix IsBad(String)Read/WritePtr loops
[KERNEL32]: SetErrorMode now handles SEM_NOALIGNMENTFAULTEXCEPT.
[KERNEL32]: Set/GetErrorMode do not set LastError on failure.
[KERNEL32]: RaiseException can simply use memcpy instead of copying argument by argument.
[KERNEL32]: Minor formatting changes.
svn path=/trunk/; revision=54291
- StackLimit was set incorrectly.
- Code was not using BaseStaticServerdata, but querying NT instead.
- Fix memory leak in failure case.
- StackCommit and StackReserved values were not aligned correctly.
- Windows Server 2003+ feature of "Guaranteed Stack Commit Size" was not respected.
- Some math was screwy.
- Failure to get NT headers was not handled.
svn path=/trunk/; revision=54285
[KERNEL32]: Fix some error cases in the helper string functions.
[KERNEL32]: Implement BasepIsRealtimeAllowed.
[KERNEL32]: Don't need RPL_MASK when setting up the CONTEXT, this now matches what Windows sets up (and also helps us validate the kernel is doing the right thing).
svn path=/trunk/; revision=54284
- Small performance improvement : Directly reference dll when getting its handle instead of just checking it's there and then load it
svn path=/trunk/; revision=54283
- Use a separate function, InstallFat12BootCodeToFloppy(), to install the floppy boot code
- Format the floppy to FAT12 before writing to it
- Booting FreeLoader from floppy works again (bug #6495)
svn path=/trunk/; revision=54269
- Prevent a deadlock on very old controllers by sending the configure and version commands before resetting the controller
svn path=/trunk/; revision=54267
- Improve the floppy controller initialization code to fix several cases where the controller could deadlock or otherwise malfunction on us
svn path=/trunk/; revision=54266
- Icon title, menu and scroll bar do not set the death bit. Menu and scroll bar set the id only on specific window messages.
svn path=/trunk/; revision=54261
- Fix function ids, setting of the death bit when in NC destroy.
- Implement get control brush and color.
- Patch by Alexander LAW, Replicate Windows behavior of WM_SETTEXT handler regarding WM_CTLCOLOR*
svn path=/trunk/; revision=54259
- Pass a placeholder trap information, indicating that we are unlocked, to MmAccessFault in MmProbeAndLockPages as temporary hack until Mm locking is fixed
- Fixes "Assertion 'GuardedMutex->Owner == KeGetCurrentThread()' failed" when faulting in pages via MmProbeAndLockPages
- Dedicated to Jim :)
svn path=/trunk/; revision=54256
- Up Dialog.c to 1.3.31, sync port and minimized differences.
- Sync Information:
Alexandre Julliard <julliard@winehq.org> : Don't overwrite the default button id when creating the dialog structure.
Michael Stefaniuc <mstefani@redhat.de>: Avoid using long. Avoid using HIWORD() on atoms/resources.
svn path=/trunk/; revision=54255
- Fix class window procedure checks, prevents misbehaving applications from calling wrong class procs.
- Fixed callout for static control color brush.
svn path=/trunk/; revision=54249
- Use SetWindowTextA/W so that it passes through defwnd to be processed for themes.
- Fix validation of the dialog window and procedure.
svn path=/trunk/; revision=54248
- Fix SetWindowTextA/W so that it passes through defwnd to be processed for themes.
- Fix desktop checks.
- Add create window flags.
svn path=/trunk/; revision=54247
- Fix a major off-by-one bug in resource conflict checking
- Respect requested alignment when checking for valid port/memory ranges
svn path=/trunk/; revision=54238
- Report legacy resources to the PnP manager via IRP_MN_FILTER_RESOURCE_REQUIREMENTS
- Report legacy access ranges in response to VideoPortGetAccessRanges
- TODO: Handle an optional resource requirements list in VideoPortGetAccessRanges
svn path=/trunk/; revision=54237
- Slot is an output parameter not an input parameter. Treat it as such.
- Don't use magic constants
- Remove a duplicate check
- Set RangeShareable for port resources based on the resource descriptor
- Initialize and set RangeShareable for memory and port access ranges instead of leaving garbage in there
svn path=/trunk/; revision=54235
- Fix ME_CharFromPoint relying on Wine's incorrect behavior of GetTextExtentExPointW. Already sent upstream(TM)
See issue #6196,5784,6037,6095,6187,6513 for more details.
svn path=/trunk/; revision=54234
Implement argument processing. Patch by Lee Schroeder.
The original patch was modified to fix the following issues:
- Indentation and coding style.
- Remove obsolete code: 1. The if-statement around the for-loop is not needed. 2. The call to wcsicmp() is already case insensitive (it is the 'i' in wcsicmp).
- Check the existence of a script file name and a timeout value before trying to process them.
- Call run_script() after the argument list has been processed. This fixes the argument order issue.
See issue #6552 for more details.
svn path=/trunk/; revision=54233
- Move playing log on sound to HandleLogon
- Don't use HKEY_CURRENT_USER for accessing user registry key. This key points to settings of user which started Winlogon process (SYSTEM) instead of logged user. Instead use HKU\SID.
See issue #5436 for more details.
svn path=/trunk/; revision=54232
- Mouse code cleanup
- Support MOUSEEVENTF_VIRTUALDESK flag in MOUSEINPUT properly
- Support MOUSE_VIRTUAL_DESKTOP flag in MOUSE_INPUT_DATA properly
- Do not ignore MK_SHIFT and MK_CONTROL flags in some mouse messages
svn path=/trunk/; revision=54227
- Don't set every loaded layout as default. Instead implement SPI_SETDEFAULTINPUTLANG. Fixes some minor problems with wrong layout loaded.
- Fix removing hotkey not associated with window (fixes winetest).
svn path=/trunk/; revision=54223
Mark the VGA driver as VgaCompatible too so it doesn't take precedence over the VBE driver in case both are installed.
See issue #6578 for more details.
svn path=/trunk/; revision=54221
- SysSetup: add a very tiny enhancement to Spanish translation (Elhoir, bug #6596).
- USetup: change encoding of Brazilian Portuguese file to CP 850.
- USetup: fix a few bugs in German and Italian translations.
svn path=/trunk/; revision=54219
- Use a Yes/No message box for "Quit the current game?" instead of Ok/Cancel. Patch by Lee Schroeder
See issue #6589 for more details.
svn path=/trunk/; revision=54218
- Fix profiles path. FIXES ICONS IN LIVECD!
- Add few interfaces to livecd registry so it doesn't spam the log so much
svn path=/trunk/; revision=54216
* Don't even create the stlport static lib. No working C++ module links to it, and until we have a proper STL support the other modules won't compile either.
svn path=/trunk/; revision=54215
- If no keyboard layout can be loaded, load US layout
- Try to load all layouts from Preloaded key even if some of them fail
svn path=/trunk/; revision=54212
- Fix path support, now all wine path tests pass.
- Need a PolyDraw test for non-path functions! The non-path code was synced ported, but still failed 16 path tests, then path was moved out. Purged missing patches from the last trunk freeze, tested with the program Area.
- Sync Information:
Evan Stade <estade@gmail.com> : Improved PolyDraw in path closed case. Added PATH_PolyDraw.
svn path=/trunk/; revision=54210
- Fix crash in PATH_ExtTextOut. See bug 6587.
- Sync Information:
Huw Davies <huw@codeweavers.com> : Add support for ETO_PDY and improve world transform support.
Massimo Del Fedele <max@veneto.com> : Fix PATH_add_outline when mapping mode != MM_TEXT. PATH_ExtTextOut remove incorrect shift to DC origin. Correctly handle space char on Path_ExtTextOut().
Dmitry Timoshkov <dmitry@codeweavers.com> : The MAT2 parameter of GetGlyphOutline is mandatory.
svn path=/trunk/; revision=54209
- Make keyboard layouts code more compatible
- Layouts and kbd files are user objects
- Preloaded layouts are added by Winlogon, not win32k
- Support Keyboard Layout substitutes in registry
svn path=/trunk/; revision=54203
- When themes are active and the text of a window changes, we should send the WM_NCUAHDRAWCAPTION message instead of drawing the window directly ( which will result painting a classic caption on the themed window)
- Fixes explorer window that is repainted with a classic style when a subfolder is opened
svn path=/trunk/; revision=54201
- Simplify monitors code
- fix few minor bugs (ie. not updating last error)
- Make monitor object more compatible to windows
svn path=/trunk/; revision=54197
- Probe input / output parameters when requests are coming from user mode
- Fix KspPinPropertyHandler to use the correct output parameter location
- Fix KsHandleSizedListQuery to use the correct output parameter location
- Tested in Vbox 4.1.4 r74291
- May fix a few sound drivers who are having user mode extra utility apps
svn path=/trunk/; revision=54191
- Apply a wallpaper and desktop background only when they have changed
- Fixes changing desktop background to the color defined by the theme
svn path=/trunk/; revision=54189
-Force all top level windows to update the size and position of their non client area when SPI_SETNONCLIENTMETRICS is used
-Should fix changing the non client metrics like caption height for open windows (before this we needed to close and reopen existing windows to let them be painted properly)
svn path=/trunk/; revision=54186
- Do not send message WM_NCCALCSIZE asynchronously because we need its return value from the receiver. Should fix resizing or moving the non-client area of windows
svn path=/trunk/; revision=54185
- Change time zone information for Russia.
- Add Russian translation of country list.
- Recode Swedish translation of USetup from CP 1252 to 850. Bug #6574.
- Fix Dutch tr. of USetup. Bug #6510, patch by <congruwer.AT.yahoo.DOT.co.DOT.uk>.
svn path=/trunk/; revision=54178
- Undefine HAVE_GETADDRINFO
- Even though this is fully implemented (and used other parts of ROS), it seems to enable broken code paths in some WINE code
- Fixes the rapps regression
svn path=/trunk/; revision=54174
- Fix TME. Not complete. These changes will allow theme highlighting to work.
- Passes wine test_TrackMouseEvent.
- Adamopoulos TME tests are proving to be difficult in locking down thread issues. On a slow system in CMD not redirected the test passes all but 15. Slow systems in CMD redirected to a output file or on real hardware in CMD not redirected, the results vary from 15 when the CMD buffer is full to 35 when it is not. When isolating each test, it passes. The QS bit issue will be handled later.
svn path=/trunk/; revision=54172
- Differentiate between incomplete and complete NCEs with regard to the timeout
- Change the IP timer to fire only once each second
- Remove unneeded completion functions
svn path=/trunk/; revision=54171
- Fix timer resolution so it will run every 10 ms +/-1ms. This is the default setting when the user provides less than 10 ms for the period to wait for the timer event. Pass wine test_timers_no_wnd.
svn path=/trunk/; revision=54170
- Do not transmit fragments recursively since it can cause a kernel stack overflow with large packets
- Big thanks to hto for his work on this bug which has eluded me since last year
See issue #5796 for more details.
svn path=/trunk/; revision=54169
- Romanian translation of userenv, winmm, ntvdm, and win32csr (bug no. 6580) and inf (bug no. 6577) by Stefan Fulea (minor corrections by me);
- Polish translation update/cleanup of inf by me;
- Spanish inetcpl translation (bug no. 6472) and French subst translation (bug no. 6166) by Jason Genie;
svn path=/trunk/; revision=54168
- Prevent corruption of the search context list using a combination of references and broader spin lock usage
- Fixes bug 6506
svn path=/trunk/; revision=54167
- Remove ReadRegistryValue in kbdlayout.c as we already implement RegQueryValue for reading from registry in W32k. This function additionally checks value type.
- Rename KBL to KL based on Techwiki
- HKCU\Keyboard Layout\Preload value is REG_SZ (not REG_EXPAND_SZ)
- NtUserLoadKeyboardLayoutEx should set last error if flags are invalid
- Minor keyboard layout code cleanup
svn path=/trunk/; revision=54163
- Reapply the ConvertThreadToFiberEx changes from r52804, they were lost in r52805. Confirmed by Alex.
See issue #6394 for more details.
svn path=/trunk/; revision=54162
- Fix insertion of the GRAPHICS_DEVICE into the global list. Fixes mode enumeration when the 2nd adapter is used.
- Remove a hack.
svn path=/trunk/; revision=54156
- Increase size of the homepage link control to correctly show the complete focus rectangle. Patch by Elton Chung (elton328 at gmail dot com)
See issue #5038 for more details.
svn path=/trunk/; revision=54154
Mark the VBE driver as VgaCompatible. This is an outrageous lie, but it lets us use 3rd party drivers, like VBoxVideo, while VBE is installed. Consider this a hack, which is to be fixed bei either making vbemp really vga compatible or by fixing whatever keeps the VBoxVideo from replacing vbemp.
svn path=/trunk/; revision=54152
- Fix a bug when iterating through the DEVMODE list
- Handle VgaCompatible flag differently, by reading this value from the registry for every installed driver.
- Priorize non-vga compatible devices over vga compatible, unless /BASEVIDEO is requested
- Fall back to vga compatible driver when no other is present
svn path=/trunk/; revision=54151
- don't define _cdecl and __cdecl to nothing on MSVC builds.
- define cdecl and CDECL to nothing
- add #pragma once
- this file is full of hacks
svn path=/trunk/; revision=54149
* Sync config.h with recent Wine and remove some now unneeded defines. By Samuel Serapion.
See issue #6174 for more details.
svn path=/trunk/; revision=54146
* Move the wctype table from kernel32 to libwine in preparations for implementing crt locale support. By Samuel Serapion.
See issue #6538 for more details.
svn path=/trunk/; revision=54144
- Properly set the timer event id's to match tests and add one more bit for track.
- Restore TME user API (now in mouse.c) and Hover Timer, these are not the Droids you are looking for.
- Restore TME cancellation point in co_MsqInsertMouseMessage and duplicate the the same in co_IntProcessMouseMessage. First thing, pDesk->spwndTrack is never NULL. The Theme TME patch code was setting the wrong QS bit and sending to the wrong queue instead of the pDesk->spwndTrack queue. Surprising this was working at all.
- Tested with Themes, passes wine test_TrackMouseEvent and fixes bug (6257 <-- !).
Notes:
It's all in the sequencing. Locate cancellation points, tracking mouse moves and set the hit test and track window. When the window message queue does not match the current thread, check it again and restore the mouse hover window if it is the same window as the track window and the hover bit set. Track cancellation points and clear the bits at the same time. Not when in WM_MOUSEMOVE. Guess when posting or sending to another thread? co_MsqInsertMouseMessage?
svn path=/trunk/; revision=54142
- Get rid of RETURN macro in input functions
- Don't check functions parameters when USER lock is held
- Improve formatting
svn path=/trunk/; revision=54141
Fix the confusion between mode index 12 and video mode 12h introduced more than six years ago. ReactOS should now boot again when VGA is selected in the 1st stage.
svn path=/trunk/; revision=54139
* Fix a copy-paste error in EngBitBlt() and improve DIB_32BPP_ColorFill() behaviour in case when input rectangle is invalid (RECTL structure fields are signed, so "jle" instruction should be used instead of "jbe").
* Patch by Igor Sudarikov <4seev3 at gmail [dot] com>
See issue #4423 for more details.
svn path=/trunk/; revision=54138
Remove an old hack, that caused an inconsistency between the display's device object name and the name of its symbolic link / registry key. The hack isn't needed anymore, because win32k now loads the next device if one isn't available. Fixes VBoxVideo being loaded together with framebuf.dll instead of VBoxDisp.dll and now the mouse pointer integration works completely.
svn path=/trunk/; revision=54135
- Keyboard state is update when msg is peeked from queue (not posted) and before virtual keys are simplified. Fixes 2 user32:input winetests.
svn path=/trunk/; revision=54134
- Since some display drivers don't do as they should - return the drisplay driver name in the DEVMODE's dmDisplayName field - but return the miniport driver name, fix that name when creating the mode list. This fixes loading of VBox driver, when it is installed properly, which needs to be fixed as well.
svn path=/trunk/; revision=54133
- Remove _fltused from alldiv_asm.s / mscmain.c and use fltused.c instead.
- Remove amd64/alldiv.S
- Implement amd64 ceil, ceilf with sse
[NTOSKNRL]
- Fix amd64 exports
[ASM]
- Fix amd64 asm files to conform to the new syntax
svn path=/trunk/; revision=54131
- Remove outdated comments
- Make Raw Input Thread CSRSS thread too
Thanks to Smiley and ThFabba for investigating TEB issue
svn path=/trunk/; revision=54128
- Added Print Screen support in previous comit accidentally...
- Fix build and warning about global object being removed
svn path=/trunk/; revision=54126
- Fix and simplify retrieving the current process window station. All processes that enter win32k have a window station (including csrss)
svn path=/trunk/; revision=54124
- Enable usbdrv.sys in 1st stage setup and the system hive
- USB keyboards and mice work now (file bugs if you have problems)
svn path=/trunk/; revision=54123
- Add a small hack to bind all keyboard port devices to the same keyboard class device when no default setting is present (circumvents win32k not supporting multiple keyboard class FDOs)
svn path=/trunk/; revision=54120
- serialui, setupapi, shimgvw and tapiui romanian translation, patch by Stefan Fulea;
See issue #6558 for more details.
svn path=/trunk/; revision=54119
- Fix two bugs revealed by the newly added test in r54114.
* Original length of the passed string should be used when checking for out-of-buffer access in the loop
* PathChars should be updated if a DOS device name was found in the given string.
See issue #6390 for more details.
svn path=/trunk/; revision=54116
Improve .proc / .endproc syntax, so that the full symbolic name is passed, which allows to create fastcall symbols. Remove the need to pass the function name to .endp
svn path=/trunk/; revision=54112
Bail out of compare_emf_bits earlier. This way we only record one skipped test, but it allows us to actually complete the metafile tests. Restore formatting to its previous glory.
svn path=/trunk/; revision=54111
- Convert wParam in WM_*CHAR messages to ANSI in GetMessageA instead of win32k. Fixes wrong characters used in cmake notepad which is ANSI.
svn path=/trunk/; revision=54109
- Right ALT on some keyboard sends LCONTROL and RALT key events
- Shift has never extended bit in key messages
- Fixes about 32 user32:input winetest
- Don't allow to register hotkeys with special keys combinations (eg. WIN key)
- Set last error in RegisterHotKey if hotkey is already registered
svn path=/trunk/; revision=54107
Allow creating of shortcuts with icon indexes != 0.
[EXPLORER]
Shuffle the resources a bit so icons with indexes 0 and 1 match those in windows explorer.
See issue #6544 for more details.
svn path=/trunk/; revision=54105
- Restore global class operations.
[ComCtl32]
- Use local class since this has higher priority over all global classes.
[Themes]
- Tested with XP Luna and two others. Rune and another Luna.
svn path=/trunk/; revision=54104
- Fix duplicated entries in screensaver selector. Patch by Carlo Bramix <carlo dot bramix at libero dot it>.
See issue #6533 for more details.
svn path=/trunk/; revision=54098
On x86 builds use the asm versions of the byte swap functions. This is neccessary for RtlUlonglongByteswap, because the function breaks the calling convention on Windows, not cleaning up the stack. Make the asm compatible to ML and improve it a bit. Investigated by Thomas Faber.
svn path=/trunk/; revision=54097
Do not use image base addresses above MmHighestUserAddress in MmMapViewOfSection, because it would first succeed when checking for conflicting memory areas, and thus not try to find a better address, but later fail in MmMapViewOfSegment, which cannot create a memory area in system space (only possible with MmKernelAddressSpace / PsIdleProcess).
See issue #5680 for more details.
svn path=/trunk/; revision=54096
- Do not keep the process handle of a started process open
- when sending a STOP command to a process failed, then succeed and mark the process as stopped.
Fixes ghost processes, when terminating a service with taskmgr and allows to restart services that got terminated or possibly crashed.
svn path=/trunk/; revision=54095
- Handle a NULL buffer in RtlIpv4AddressToStringA as Windows does. Fixes crash in ntdll:rtl test
[CRT]
- Fail on qsort with 0 size. Fixes hang in ntdll:string test
svn path=/trunk/; revision=54093
* Use the gcc toolchain-provided STL.
* Use STLport only on MSVC builds.
[KSPROXY]
* Set imagebase, properly set its entry point, use set_cpp() instead of hardcoded definition/includes and add no exceptions/rtti gcc flags.
[GLU32]
* Use set_cpp() instead of the hardcoded target_link_libraries().
[PORTCLS]
* Use set_module_type() with proper entry point, instead of expanded macro calls, and get rid of hardcoded target_link_libraries().
* Commit Dedicated to Olaf Siejka (Caemyr).
svn path=/trunk/; revision=54082
Implement SystemTimeAdjustmentInformation case in NtSetSystemInformation. This silences the annoying messages caused by VBoxService
svn path=/trunk/; revision=54081
Implement support for hotkeys for run, show desktop and logoff
currently it also causes the startmenu to open, like with win-E, but this is a different bug
See issue #1512 for more details.
svn path=/trunk/; revision=54076
- Properly copy image name in TH32CreateSnapshotSectionInitialize. Fixes truncated file names in kernel32:toolhelp test
svn path=/trunk/; revision=54071
- Hackfix RtlGenRandom to be slightly less dependent on the current time, while staying very un-random.
- Should fix services problems (no joke) due to non-unique RPC UUIDs
See issue #6542 for more details.
svn path=/trunk/; revision=54069
- Use 64 byte bitmap for keystate instead of 256 elements array (like Windows)
- Fix GetAsyncKeyState lowest bit meaning. See MSDN
- Call low level keyboard hook before setting async key state for keys which distinguish between left and right hand. Fixes user32:GetKeyState apitest failure
- Improve IsHotKey and delete GetHotKey function to avoid code duplication
- Support WIN key in hotkeys code instead of keyboard.c
svn path=/trunk/; revision=54063
set CF, ZF and SF based on REGS structure, when doing int's. Fixes slowness during 2nd stage boot in QEMU. Dedicated to Samuel.
svn path=/trunk/; revision=54062
- SetupDiCreateDeviceInfoW: return the proper error code. Fixes a winetest.
- Implements parts of SetupDiRegisterDeviceInfo.
svn path=/trunk/; revision=54059
- Implement proper security cookie initialization code (same algorithm as in sdk/crt, however using as much data directly available as possible).
- Don't initialize security cookie as a 32bit value when 16bit security cookie was requested. This fixes msi-related crashes (or any other executables which has 16bit security cookie set).
See issue #6341 for more details.
svn path=/trunk/; revision=54057
* Temporarily go back to using the ret^H^H^H previous way of using debug info. Build will, for example, be slower, take double the disk space it needs right now, but...
* Introduce __ROS_DWARF__ for the meaning __ROS_CMAKE__ implied previously.
svn path=/trunk/; revision=54055
- Fix possible integer overflow in RtlpInitializeHeapSegment
- Simplify loop in RtlIsDosDeviceName_Ustr
- Make MonthLengths an array of UCHARs instead of ints
- Remove pointless loops in RtlTimeToTimeFields
- Fix MSVC warnings
svn path=/trunk/; revision=54054
* Use the previous registration mechanism. Fixes several shell32 regressions.
* Dedicated to Claudiu Mihail and Johannes Anderwald.
svn path=/trunk/; revision=54053
- Do not call UserSetCursor in MsqCleanupMessageQueue because it uses current thread queue instead of queue in cleanup (they can be different).
See issue #6530 for more details.
svn path=/trunk/; revision=54051
- Rename KTHREAD.DispatchHeader to Header
- Remove KeInitializeDispatcherHeader and initialize dispatcher headers depending on object type. Fixes all ntos:KeEvent and KeTimer kmtests.
svn path=/trunk/; revision=54048
- Fix symbolic link creation with a reference string
- Audio devices start successfully again (sound still seems to be regressed however)
svn path=/trunk/; revision=54040
- Do not fail in SetupDiInstallDevice if the inf file already exists. Fixes VirtualBox 4.1.x additions
See issue #6522 for more details.
svn path=/trunk/; revision=54037
Fix memory leaks
Fix broken if condition
Use delete[] operator for allocations made with new[] operator
Properly check for failed allocations by not throwing exception
svn path=/trunk/; revision=54034
- Save special values in F/Blink when Ex(f)InterlockedRemoveHeadList is called on checked build. Fixes all ntos:ExDoubleList kmtests
svn path=/trunk/; revision=54026
- Add a couple of tests for the messages sent when changing the z-oder of windows with DeferWindowPos (this test ignores owners z-order)
svn path=/trunk/; revision=54025
- remove all previous SetActiveWindow tests as they were completely wrong. This is because changing the z-order of the owner windows is done by SetWindowPos as tested in the same test suite. However SetActiveWindow calls SetWindowPos without SWP_NOOWNERZORDER parameter which leads to changing the zorder of owner windows
svn path=/trunk/; revision=54023
- Add some more tests for SetWindowPos with or without SWP_NOOWNERZORDER. This clears the misunderstanding on changing owner window z-order once and for all
svn path=/trunk/; revision=54022
Fix handling of RLE compressed bitmaps in NtGdiStretchDIBitsInternal, by using similar code as in NtGdiSetDIBitsToDeviceInternal calling GreCreateBitmapEx instead of NtGdiCreateDIBitmapInternal.
Fixes bug 6168.
svn path=/trunk/; revision=54019
- Post-r52687 implementation of RtlIsDosDeviceName_U is more advanced than pre-r52687 one and covers more corner cases which may not be tested by existing winetests.
- Thus bring back the post-r52687 and also introduce fixes for the following bugs in it:
* Incorrect extension/space-at-the-end truncation
* Incorrect encoding of returned offset/length pair
* Incorrect condition for for-loop exit
* Incorrect calculation of the return length in case of COMx/AUX devices (it should not include the terminating 0 character)
- After those trivial fixes (mostly oneliners and typos), the function works as expected and passes all ntdll_winetest path tests. Sorry for any inconviniences!
svn path=/trunk/; revision=54018
- Implement support for scatter/gather DMA
- Tested with Intel's NDIS 5.1 PRO/1000 driver (but more testing would be helpful to weed out bugs)
[NDIS]
- Report NDIS 5.1 now since S/G DMA is supported
- Stop warning when S/G DMA is used
svn path=/trunk/; revision=54016
- Do not treat all extended keys as power keys in HandlePowerKeys. Fixes WM_KEYDOWN not being send right after extended key has been pressed.
See issue #3212 for more details.
svn path=/trunk/; revision=54007
- Rewrite keyboard input processing to match windows behavior
- Deduplicate code, add comments, mark functions static
- Support keyboard LEDs in SendInput API
- Set last error in more places
- Make keyboard layouts compatible with Windows
- Add some defines from keyboard layouts to kbd.h and use names from MS file
- Move kbd.h from ntoskrnl to NDK as it has nothing to do with kernel
- More fixes which I don't remember
- Please test CAPSLOCK key with AltGr characters (RALT+X) for layouts which support it. If something fails, report it to me on IRC
svn path=/trunk/; revision=54006
- Restore a ros-diff lost during wine sync. Should fix kernel32:volume crash
- Better mark ros diffs in testlist.c and also bring it somewhat into sync
svn path=/trunk/; revision=54003
- netid, vmwinst romanian translation, see bug number:6531;
- resource updates and corrections in multiple romanian translations, see bug number: 6532;
- Patches by Stefan Fulea;
[ACLUI]
- convert module to ReactOS resource style (separate lang directory);
svn path=/trunk/; revision=53998
Improve the new heap to use a freelist, which boosts allocation performance by a factor of 7. Its now even slightly faster then bget.
svn path=/trunk/; revision=53989
Fix a bug, when the memory for a file cluster chain record was freed from the heap although it was still in use. Instead free the buffer when the file is closed.
svn path=/trunk/; revision=53982
- Do not access an object after destruction.
- Remove the remaining mentions of a function which Sir Richard defeated a year ago.
svn path=/trunk/; revision=53978
- Add a flag to distinguish Ansi and Unicode message tables. Keep them Ansi for now pending further testing
- Apply this to MSVC build as well. Fixes ntoskrnl messages incorrectly being Unicode
svn path=/trunk/; revision=53971
- Clean up some headers
- Add DPRINT_HEAP debug channel mask
- Make ExAllocatePool a function instead of a macro
- Add a cleanup function to the new heap code
svn path=/trunk/; revision=53967
- Implement CM_Get_DevNode_Custom_Property[_Ex]A/W.
- Fix the return value of CM_Open_Class_Key_ExA.
- Handle REG_MULTI_SZ values properly in CM_Get_DevNode_Registry_Property_ExA.
svn path=/trunk/; revision=53965
- Correctly handle an arbitrarily large sockaddr in WSPConnect. Fixes mIRC 7.1x crash on connect
See issue #6005 for more details.
svn path=/trunk/; revision=53962
* Work around an issue with clang and __debugbreak(). They mark __debugbreak as built-in (so we can't use ours) but it's not supported yet (so if we use nothing we get an error).
svn path=/trunk/; revision=53958
Implement new heap code. The code is much less complex than bget and allows to use allocation tags and releasing unused memory. Its not yet active.
svn path=/trunk/; revision=53956
- Pass the module list head to a number of pe loader functions instead of the loader block.
- use static for some functions
- remove unneeded prototypes
svn path=/trunk/; revision=53947
- Fix RECTL_bPointInRect to work like PtInRect
- Use the fixed RECTL_bPointInRect in DIB_XXBPP_FloodFillSolid to fix a bugchek
See issue #6437 for more details.
svn path=/trunk/; revision=53946
* Reduce the scope of a variable.
* Comment out some unused code.
* Initialize some member variables in the respective constructors.
svn path=/trunk/; revision=53944
- use RtlStringCbPrintfA instead of manually calculating required length and checking if the buffer is large enough
- Use %wZ as format specifier for a UNICODE_STRING, instead of using %S and making assumptions about zero termination.
- Don't "while (TRUE);" on buffer overflow!
svn path=/trunk/; revision=53942
Fix remaining issues, bootsector compiles with GAS and ML. Sadly GAs sucks a bit and does neither allow to specify byte offsets, nor does it chose this itself to save space. As a result the code gets several bytes larger and I had to trim the messages even more.
svn path=/trunk/; revision=53938
- Somehow our shell now expects the icon of My Computer to be the resource icon %SystemRoot%\Explorer.exe,0. Change the resource id of My computer icon to 0 to fix the wrong icon on the desktop
svn path=/trunk/; revision=53927
* Add a new Starfield screensaver. By Carlo Bramini (carlo DOT bramix AT libero DOT it).
* Adapted the build system files to get it to compile with current trunk.
See issue #6270 for more details.
svn path=/trunk/; revision=53920
use a dll instead of an exe for the freeldr_pe and setupldr_pe targets and add a spec file to export scsiport stuff. This is the first step in cleaning up the scsiport import hack in freeldr
svn path=/trunk/; revision=53914
- Enable all interrupts at the same time to avoid a condition where the controller stops responding to KBD_READ_MODE on some systems
- See bug #6271 for details
svn path=/trunk/; revision=53911
- Implement MiInitializePfnAndMakePteValid() which is basically MiInitializePfn() combined with making PTE valid and some other changes.
- Don't initialize PTE and pass valid PTE for initialization to MiInitializePfn() when allocating a page in special pool, this is wrong. Instead call a function designed to do this - MiInitializePfnAndMakePteValid(). It performs correct checks and PFN reference counting.
svn path=/trunk/; revision=53909
Remove DTEs for ntbootdd.sys and freeldr.sys after they are not used any longer. freeldr one will even point to invalid stack data. It only works currently, because the LoadOrderListHead is reset later. More fixes are coming.
svn path=/trunk/; revision=53908
- properly cleanup when tests are completed
- Add some more tests for SetActiveWindow that prove that bringing owner/owned windows to front is done by SetActiveWindow and not by SetWindowPos
svn path=/trunk/; revision=53905
- Implement SHPropStgCreate, SHPropStgReadMultiple, SHPropStgWriteMultiple based on wine
- This should fix the shellole tests
svn path=/trunk/; revision=53903
- implement helper functions MiIsMemoryTypeFree and MiIsMemoryTypeInvisible
- use symbolic names instead of hex values
- Don't ASSERT on mising large page support, just warn
- add some more mm macros for amd64
svn path=/trunk/; revision=53902
- Report the device's lockable status in the device capabilities request
- Only report the device as disabled if it is both not enabled and not functional
svn path=/trunk/; revision=53899
- Move some disk related stuff that is unrelated to the registry data into a new file, hwdisk.c
- Don't get the disk count from the size value of a structure that was previously calculated from the disk count, but instead save it in a global variable.
- Initialize certain data in a better place
svn path=/trunk/; revision=53896
- Update ACPICA from 20091214 to 20110922
- Rewrite a large portion of the OSL code
- Perform a full initialization of ACPI objects (don't skip devices and events)
- Please retest ACPI bugs after this commit
svn path=/trunk/; revision=53895
- Add a hack to skip calling methods of the HPET object because it causes a hardlock on VMWare
- Attempt #2 at fixing VMware
svn path=/trunk/; revision=53892
- Do not force the hard error dialog to appear. As a result, calling SetErrorMode now really prevents the hard error dialog from appearing.
svn path=/trunk/; revision=53889
- Minor fixes to ScmAssignNewTag
- Correctly handle an invalid parameter case in RCreateServiceW. Fixes second stage boot under certain circumstances.
svn path=/trunk/; revision=53886
- Remove pointless NTKRNLAPI
- remove unneeded include
- delete deprecated file amd64/loader.c
- add an #ifef __REACTOS__ around public header includes to make it possible to use an alternative header set from ntsup.h
svn path=/trunk/; revision=53885
- The width parameter in AcpiOsReadPciConfiguration and AcpiOsWritePciConfiguration was in bits but we were treating it as a width in bytes
- This caused overreads, memory corruption, and crashes when these functions were called (VMWare was particularly picky about bad accesses to the PCI configuration space)
- A hack was (unknowingly) added which prevented some crashes but had a side-effect of causing the partial disruption of ACPI's PCI configuration space accesses while the others that went through wrote bad data to the PCI config space or corrupted kernel memory
svn path=/trunk/; revision=53880
- Modify some memory types to match what windows uses
- Modify PcMemGetBiosMemoryMap, so that is crops free memory to page alignment, while extending non-free memory to page alignment. This way we don't loose small firmware descriptors like the one for the extended bios data area enumerated by the bios. Fixes a warning from win 2003 about not owned memory.
- Modify MempAddMemoryBlock to map LoaderFirmwarePermanent into kernel space like ntldr, but don't map page 0! ntldr doesn't do that either. Also map LoaderXIPRom.
- After generating memory descriptors from the page lookup table, add any descriptor from the bios memory table, that is higher than MmHigestPhysical address, so we don't loose high rom descriptors. Do not map beyond LoaderPagesSpanned.
svn path=/trunk/; revision=53878
- copy ebp into REGS structure on Int386
- Write pointer and size of extended bios data area in physical page 0 location 0x740 and pointers to vga rom fonts in location 0x700. This is required by windows 2003's videoprt. Now freeldr is more Windows 2003 ready then ever!
svn path=/trunk/; revision=53876
- Fix auto complete init method based on Wine. Autocomplete shell tests should no longer crash now.
- Brought to you by Claudiu Mihail.
svn path=/trunk/; revision=53871
* Implement CDefView::SetCurrentViewMode based on Wine, along with some other bugfixes. Shell32 shlview winetest doesn't crash anymore as a result.
* Brought to you by Claudiu Mihail.
svn path=/trunk/; revision=53869
- Introduce two MM_EMPTY_LIST definitions - one for PTE as a whole and one for MMPTE_LIST NextEntry field only and move them to the actual architecture specific headers.
- Implement special pool. This is a special (Capt. Obvious!) kernel pool which is used to catch overruns or underruns, freed memory access and perform additional validation for paged/nonpaged allocations access.
svn path=/trunk/; revision=53868
- Do not subclass user32 controls when themes are disabled.This is a temporary hack bacause it turns out that subclassing causes problems. When it is fixed this will be reverted. This change means that in order to make themes work properly, the user has to reboot after enabling.
svn path=/trunk/; revision=53866
- Fix memory type for hal and referenced dlls
- Don't allocate the LPB, MADs and all the other stuff from the heap, instead allocate one block of memory for the LPB, extension, etc and strings (fits into one page) and another page for the MADs limit to 200 MADs, 1024 is way over the top)
- Map only those pages into kernel mode, that the kernel will need, this way we don't need to unmap anything later.
svn path=/trunk/; revision=53864
- Fix a typo in HalpAllocPhysicalMemory, that caused the function to remove MADs that still had pages rather than removing those who are empty. Fixes an assertion on VMWare.
Kudos go to Kamil for tracking it down.
- Fix another bug in HalpAllocPhysicalMemory, where the size of the newly allocated MAD was set to the alignment value instead of the original MAD, this lead to conflicting MADs and possible reuse of hal memory by the kernel. This seems to fix a bugcheck 0x19 with halacpi.
svn path=/trunk/; revision=53861
- Get rid of MmAllocateMemory, freeing about 1MB of low physical memory for the kernel
- verify that MmMarkPagesInLookupTable is not called with invalid page regions
- add maximum allocation to heap statistics
<long explanation>
There was a function called MmAllocateMemory, what was commented as "// Temporary forwarder..." since January 2008. This function allocated one page of memory and marked it as LoaderOsLoaderHeap. This function was used by the fat filesystem code (and linuxboot.c) which allocated and freed memory in small chunks all the time. Since MmFreeMemory() is not implemented at all (obviously someone removed it) we were allocating one full page for allocations as small as 8 bytes and never free them. This accumulated to a total of 240 pages, almost 1MB, split into into 14 chunks. This memory was never freed by the kernel (the kernel keeps the loader heap memory for some reason) and fragmented the low memory region.
Remove MmAllocateMemory completely and replace references in the fat code with MmHeapAlloc. The maximum heap usage after this was 184 KB, heap size is 4MB.
</long explanation>
svn path=/trunk/; revision=53857
- In TrackMouseEvent tests, use mouse_event function to move the mouse cause this causes low level hooks to be called and SetCursorPos doesn't
- Test low level mouse hooks for TrackMouseEvent
- Test ways to block WM_MOUSEHOVER and WM_MOUSELEAVE. This is useful to detect where each message is generated
- clean up the code a bit by adding FLUSH_MESSAGES macro that checks for expected message queue status and flushes the message queue
svn path=/trunk/; revision=53855
- Improve tracing
- Fix memory corruption caused by marking pages outside of page lookup table (when table is allocated only free pages was counted but MmMarkPagesInLookupTable was used for all types of memory like firmware). Fixes bugcheck in ACPI HAL which happened for me
svn path=/trunk/; revision=53848
- Undo disabling of systemcalls, which was disabled in r53646 for testing purpose.
- Sillence "Support for SYSENTER detected." debug print, since its the normal case, only keep the disabled and not present cases.
- Silence "xxx support detected but not yet taken advantage of" debug prints. CPU features are already reported and there is no point in printing every single feature we do not support.
- Invert if/else in KiRestoreFastSyscallReturnState
svn path=/trunk/; revision=53846
Add FREELDR_MEMORY_DESCRIPTOR structure using TYPE_OF_MEMORY instead of
MEMORY_TYPE. Use it in place of MEMORY_DESCRIPTOR. Undo r53837 changes.
svn path=/trunk/; revision=53841
* Add _dbg versions of freeldr_pe and setupldr_pe per Dmitry's request. They retain debug info, and they're compiled on-demand (not included in the normal build).
svn path=/trunk/; revision=53840
Fix wrong behavior of the "native applications startup at boot time" feature in SMSS.
Modifications:
- Added missing buffer allocation checks.
- Check for presence of default path.
- Minor clean-ups.
- Fix bad indentation/coding style.
Original patch by Hermès BÉLUSCA.
See issue #6180 for more details.
svn path=/trunk/; revision=53839
- Cleanup amd64 mm.h header
- Call MiScanMemoryDescriptors earlier, since it doesn't depend on anything except the loader block.
- MiComputeColorInformation() and the calculation of the pfn database size are portable, so move them out of MiInitMachineDependent
svn path=/trunk/; revision=53838
- Implement logging post messages or winevents
- For SetActiveWindow test, also test the parameters of WM_WINDOWPOSCHANGING and WM_WINDOWPOSCHANGED messages
- For TrackMouseEvent test, also test the WM_SYSTIMER that is possible to be logged now
svn path=/trunk/; revision=53830
The following modifications were made:
- Removed the help texts because they were copied from Windows Diskpart.
- Used a single string resource per help text.
- Used only Unicode strings and Unicode string functions.
- Replaced the giant if-statement in the command dispatcher in interpreter.c by a command-table and a little loop.
- Improved the command-line parser. Argument count and argument vector are passed to the command and help functions for easy evaluation.
- Moved the help function for commands into the command specific file.
TODO:
- Implement all commands as the utility is only an empty frame.
- Add help texts.
- Translate help texts.
See issue #6402 for more details.
svn path=/trunk/; revision=53826
- Session space address layout is architecture specific, move its initialization into MiInitializeSesseionSpaceLayout() in architecture specific file
- Use dedicated constants for setting MmSystemRangeStart, MmUserProbeAddress and MmHighestUserAddress isnetad of making assumptions.
svn path=/trunk/; revision=53825
Add MiScanMemoryDescriptors, that combines the work of MiPagesInLoaderBlock, the loop in MmInitializeMemoryLimits and the loop in MiInitMachineDependent.
svn path=/trunk/; revision=53822
* Bring-in some idl changes from Wine. These changes (and the related changes below) are to allow us to sync shell32 tests to Wine 1.3.29.
[UUID]
* CLSID_TaskbarList is no longer needed in here.
[ACTXPRXY]
* Sync with Wine 1.3.29.
svn path=/trunk/; revision=53820
Converting MmLargeStackSize from registry setting to byte value is portable code, move it out of MiInitMachineDependent
svn path=/trunk/; revision=53819
- Use TYPE_OF_MEMORY from the beginning instead of MEMORY_TYPE and later converting it.
- Disable tracing, accidentally comitted in previous revision
svn path=/trunk/; revision=53818
Fix some bugs in the code that deals with memory descriptors:
ArcGetMemoryDescriptor was enumerating the bios generated entries first and then the static entries. The latter were conflicting with the former and took precedence when initializing the page lookup table. The problem was, that MmFindLocationForPageLookupTable would use the highest range of pages available. If that range would be conflicting with a following static descriptor, the static would have been ignored. It only worked because on x86 the highest bios descriptor has enough free pages, before it conflicts with a static entry (page 0xfff, marked as unusable), so that the page lookup table could be created.
MmGetAddressablePageCountIncludingHoles enumerated all memory descriptors to find MmLowestPhysicalPage, but it was only counting FreeMemory, thus skipping all other memory ranges. This only worked, due to the previous bug, so that the bios pages shown first took precedence over the following static descriptors. Without the former bug MmLowestPhysicalPage would be 0x100 which would tigger in the next bug:
MmAreMemoryPagesAvailable took the passed address range and looked up all pages in the lookup table to see whether all are free. Now the function didn't check, whether the passed address was below MmLowestPhysicalPage and in case it was, happily accessed the memory below the lookup table. This would result in hal being loaded at 0x40000 overwriting the loader itself.
This is all fixe by implementing a new way of creating the memory map. First there is a static memory map, which has enough free entries to add dynamic ranges. Then AddMemoryDescriptor will allow you to add a new range,
while the already existing ranges will take precedence and the new ranges will be properly split and inserted, so that the resulting map does not contain any overlapping regions and is sorted from low to high pages. The static memory descriptor exists in the architecture specific file.
The code that enumerates the bios information now uses this function to add the ranges on top of the static descriptor.
More cleanup work to follow.
svn path=/trunk/; revision=53816
- Move shared code used to log window messages in helper.c
- Implement logging hooks
- Fix comparing the logged messages to show the correct line of the test
- Add tests for hooks in TrackMouseEvent test
svn path=/trunk/; revision=53812
- Fix several typos. Patch by Elton Chung (elton328 at gmail dot com). Slight compatibility fix by me
- Add missing NTSYSAPI to some externs
- remove Se*DefaultDacl externs from ntifs, as they don't belong there, and are already correctly found in ndk/setypes.h
svn path=/trunk/; revision=53809
- Revert part of r53788 (MAXIMUM_IDTVECTOR is a valid vector number)
- The buffer overrun in the IDT code was already fixed by r53420
svn path=/trunk/; revision=53806
- Multimodule translation and UTF-8 conversion to Portugese Brazilian. Patch by mkbu95. Minor fixes by me;
See issue #6467 for more details.
svn path=/trunk/; revision=53805
- Fix a bug, where the bios memory map could have a bogus entry: according to R.B.'s Interrupt list, in some bioses the function for enumerating the memory can return ebx != 0 for the last descriptor, and fail in the following call returning CF = 1. In that case we were previously counting that last descriptor as well, but its content wouldn't describe a valid memory region.
- Use a static buffer for the bios memory map, stored in the achitecture specific file and return a pointer to that variable from PcMemGetMemoryMap. Reuse the static map from DetectAcpiBios instead of enumerating again. On arm, we don't even need a buffer, we return the pointer from the arm block.
svn path=/trunk/; revision=53803
Patch by Brian Palmer:
Fix problems that could occur, when the BIOS returns memory regions that are not page aligned, by fixing up the bios memory map before using it.
svn path=/trunk/; revision=53797
- Implement the LSA object database.
- Implement the information classes PolicyPrimaryDomainInformation and PolicyAccountDomainInformation of LsarSetInformationPolicy().
svn path=/trunk/; revision=53783
- Keep APC delivery disabled while holding a Heap Lock, not just while acquiring/releasing it (Dedicated to Timo and Ged).
- Store the Heap Lock in non-paged pool, the only place where executive resources belong (Dedicated to patient people).
[RTL]
- Restructure/simplify/correct Heap and Heap Segment initialization (partially to cope with the changed Heap Lock interface).
- Restrict the location of Heap Segment headers to the base address of the Heap Segment (which frees up a whopping 60 bytes per Heap!).
- Cater for acquiring the Heap Lock exclusively or shared (the latter is only available in kernel-mode); only exclusive locking is used for now.
- Use a plain critical section to guard the Process Heap list, no reason to disguise it as a Heap Lock (saves us a handful more bytes).
svn path=/trunk/; revision=53761
Fix CsrProbeForRead and CsrProbeForWrite to actually access the memory, by using volatile. The compiler optimized the access away previously.
svn path=/trunk/; revision=53760
- Fix a number of MSVC/amd64 warnings/problems
- Remove code duplication in RtlCharToInteger and RtlUnicodeStringToInteger
- Fixup too large string length in RtlInitAnsiString, RtlInitUnicodeString
- Use symbolic constants instead of hardcoded values
- Fix overflow check in RtlInitUnicodeStringEx
svn path=/trunk/; revision=53746
- Merge changes from the themes branch. These changes will introduce test regressions for class functions and for TrackMouseEvent. The changes in the class functions are needed in order to let comctl32 register its own versions of built in controls. This is a hack and will be removed when we properly support class versions. The second change that causes test regressions is in TrackMouseEvent and it is needed in order to fix WM_MOUSELEAVE messages that were completely broken and not tested at all by wine's test suite.
svn path=/trunk/; revision=53745
- Make Vector parameter in HalEnableSystemInterrupt, HalDisableSystemInterrupt and HalBeginSystemInterrupt an ULONG, not an UCHAR
[NDK]
- 64bit fixes for HANDLE_TABLE, KPROCESS, SECTION_IMAGE_INFORMATION, MMADDRESS_LIST, MMVAD_FLAGS, MMVAD, MMVAD_LONG, MMVAD_SHORT, MEMORY_DESCRIPTOR, MEMORY_ALLOCATION_DESCRIPTOR, LdrVerifyMappedImageMatchesChecksum
- KDPC_DATA::DpcQueueDepth is signed on amd64, unsigned on x86
[NTOSKRNL]
- Fix hundreds of MSVC and amd64 warnings
- add a pragma message to FstubFixupEfiPartition, since it looks broken
- Move portable Ke constants from <arch>/cpu.c to krnlinit.c
- Fixed a bug in amd64 KiGeneralProtectionFaultHandler
svn path=/trunk/; revision=53734
Fix the overly complicated and buggy implementation of RtlIsDosDeviceName_Ustr based on the simple pre-r52687 implementation of RtlIsDosDeviceName_U. This fixes 14(!) wine tests.
svn path=/trunk/; revision=53726
RtlGenerate8dot3Name: Ignore spaces in long file names instead of converting them to underscores.
See issue #6385 for more details.
svn path=/trunk/; revision=53725
- Don't check for leading "." char in a filename, because it's valid in case of LFN. The driver is already quite broken, so it can't get more broken if this assert is commented out.
Fixes assertion in "kernel32_winetest module".
svn path=/trunk/; revision=53722
- Fix KiGeneralProtectionFault
- Don't bugcheck in KiUnexpectedInterrupt
- Remove debugprint in KiEndInterrupt, and instead assert, its called from the clock interrupt, since thats the only instance we handle right now
svn path=/trunk/; revision=53720
- add common amd64 asm trap macros
- fix amd64 version of HackEoi
- send EOI in amd64 trap exit code, so we avoid using HackEoi
- modify HalpClockInterruptHandler for better portability
- fix amd64 TscCalibrationISR
svn path=/trunk/; revision=53719
- get version.h and buildno.h out of precompiled headers, to avoid rebuilding the whole module each time we reconfigure the build
svn path=/trunk/; revision=53716
- Fix some stuff that gcc complains about. It doesn't like to use static data in non-static inline functions for some unknown reason.
- Implement support for level triggered interrupts, when using lazy irql
svn path=/trunk/; revision=53711
- Improve List_PaToVa, making the code cleaner
- Dereference LoaderBlock->Extension only after WinLdrSetProcessorContext, since its in kernel address space
svn path=/trunk/; revision=53710
- Move some shared constants (HYPERSPACE_BASE, MM_HAL_VA_START, APIC_BASE) to NDK to avoid duplicating them, also fix wrong values for amd64
- Map the hal address space in amd64 freeldr
- Fix use of uninitialized variable (HighPart of PHYSICAL_ADDRESS) in HalpAcpiCopyBiosTable
- Fix some MSVC and amd64 warnings
svn path=/trunk/; revision=53709
* Merge r53700. Point 3 in bug 6481 was fixed by fixing the parser (at line 414) to use only ',' as the delimiter, to match Windows behavior.
svn path=/trunk/; revision=53702
Implement lazy irql for APIC. This is useful for VMs, since APIC usually has high overhead due to the need of invoking the hypervisor on every irql raise and lower. With lazy irql we avoid that until absolutely neccessary. Note that we misuse the PCR's IRR field to save the current hardware irql. Its a huge performance boost (some parts take half the time), making APIC performance close to PIC performance on VBox. This is something that Windows doesn't have :)
svn path=/trunk/; revision=53695
- Create a volatile device tree during 1st stage so the new HAL that's loaded in 2nd stage can do proper enumeration without phantom devices from the old HAL
svn path=/trunk/; revision=53694
- Fix incorrect loop boundaries in the ARM version of MmCreateProcessAddressSpace. Spotted by Vincenzo Cotugno and cppcheck
- Correct outdated wineimport.cmd URL in README.WINE
- Assorted spelling and readability fixes
svn path=/trunk/; revision=53688
- Add support for guarded allocations, that can be used to detect buffer overruns
- Add GuardedMemory example test
- Add MmSection test with some basic parameter checks for MmCreateSection
- Fix some x64 build problems
- Update readme
svn path=/trunk/; revision=53687
- Add optional asynchronous io code for service control pipes. This is disabled by default due to bugs in NPFS.
- Read service pipe timeout value from the registry.
svn path=/trunk/; revision=53686
- End the interrupt in HalpApcInterruptHandler and HalpDispatchInterruptHandler before calling the kernel, otherwise we would be stuck at high processor irql
- Improve HalEnableSystemInterrupt
- disable interrupts in HalpInitializeClock
=> APIC hal boots to desktop!
svn path=/trunk/; revision=53685
The handlers for unexpected interrupts that we generate from asm macros use a push instruction with an 8 bit operand to push the interrupt number on the stack. Now the 8 bit push has the unfortunate behaviour of sign extending to 32 bit. But since the assembler is smart (or at least thinks so) it will not generate such a push instruction for values larger than 0x7f. This resulted in 7 bytes long stubs in the range of 0x30 .. 0x7f and 10 bytes long stubs after that. To fix this 128 is substracted from the value and later readded in the common code path. A second issue arose, because the assembler would start to use 8 bit relative short jumps as soon as the target was less than 127 bytes away, which happened for some of the higher interrupts, which then had a smaller stub then the others. Fix this by moving the stubs up in the code further away from the target label, so that always 32bit relative jumps will be used. These problems didn't show up so far, since we used the PIC and interrupts in the range of 0x30 .. 0x40 and the code that relied on the stubs all having the same length was working well. This changes with an APIC based hal, where interrupts of much higher number are likely to be used.
svn path=/trunk/; revision=53679
- Use a global variable HalpBuildType instead of a define when checking the BUILD (we don't recompile this code for SMP hal)
- Move registering IRQ2 out of the generic init code into HalInitializePICs
- Modify legacy HalpGetRootInterruptVector to use HalpIrqToVector and HalpVectorToIrql so it works with APIC, too
svn path=/trunk/; revision=53675
- Implement HalpApcInterruptHandler, fix HalpDispatchInterruptHandler
- Insert a read to the APIC version register in ApicGetCurrentIrql(). This fixes inconsistencies between the internal APIC state and register reads/writes.
- Remove old hacks in ApicGetCurrentIrql() and ApicSetCurrentIrql
- Fix HalpVectorToIrql()
- read/write cr8 for the value of the TPR on amd64 builds
- Fix amd64 build
svn path=/trunk/; revision=53673
- Skip the remaining tests that result in assertions
- Hide KeProcessor test, as it takes very long
svn path=/branches/GSoC_2011/KMTestSuite/; revision=53669
Add a hack to work around a bug in VBox: The APIC emulation requires an iret instruction following closely after writing the EOI register. Since we return from kernel mode traps with a jmp (yes you can return from an interrupt with a jmp!) the EOI is never triggered, making VBox believe we are still serving the interrupt and keeping the PPR on high level, preventing following interrupts. A small asm stub now does the work of both writing the EOI and doing an iret.
svn path=/trunk/; revision=53665
- Add a hack to ApicGetCurrentIrql and ApicSetCurrentIrql to work around a VBox bug
- call KeSetTimeIncrement after initializing the rtc clock
- calculate TSC frequency from the samples
- Fix a bug in KeStallExecutionProcessor
svn path=/trunk/; revision=53658
- Rename gQueueKeyStateTable to gKeyStateTable as it has nothing to do with msg queues. They have their own table
- Use KS_*_BIT defines instead of magic values
svn path=/trunk/; revision=53655
* Reintegrate the c++ shell32 branch. Exemplary team-work.. kudos !
* Better code quality, more tests run with less failures... and more.
* Dedicated to everyone who helped ;)
svn path=/trunk/; revision=53653
[FORMATTING]
- Second wave of formatting. At least now we won't have any tabs and spaces mixed in.
svn path=/branches/shell32_new-bringup/; revision=53651
- Fix bug 6336 of shell32_new. The bug was caused by not properly verifying the return value of ShellExecuteExW.
[FORMATTING]
- First wave of accidental formatting (sorry again about mixing code and formatting heh).
svn path=/branches/shell32_new-bringup/; revision=53650
- Start a new device manager to test the reactos ATL code.
- Also, the old one sucks big time and was quickly hacked together.
svn path=/trunk/; revision=53648
- Rewrite clipboard to match Windows implementation
- Clipboard uses window stations instead of global variables
- Data is saved in clipboard data user objects
- Memory is no longer leaked when winsta is destroyed
- Data is synthesized on demand
- Make internal functions static
- Fix possible memory corruption in IntGetAtomName
- More winetests are passed
svn path=/trunk/; revision=53644
- Simplify accelerators code a bit
- Set last error in NtUserCopyAcceleratorTable
- Fix user32:resource winetest regression
svn path=/trunk/; revision=53642
- Make sure the ParseDisplayName functions make the ppidle parameter NULL whenever that's possible. Scores a wine test. Spotted by Victor Martinez.
svn path=/branches/shell32_new-bringup/; revision=53641
- Remove kmtest_old and kmtloader as all their functionality is included in the new test suite
svn path=/branches/GSoC_2011/KMTestSuite/; revision=53639
- Fix I/O APIC register access
- set APIC logical id based on Cpu (currently flat model with up to 8 cpus supported)
- In HalpInitializeTsc, setup the RTC clock, since the timer is initialized later
- in the TSC calibration ISR, send EOI and read RTC register C to get the next interrupt
svn path=/trunk/; revision=53634
- Initialize uninitialized variables.
- Add code to guard against potential NULL pointer dereferencing.
Thanks to Amine Khaldi for pointing out all these.
svn path=/branches/shell32_new-bringup/; revision=53633
- Don't use debug prints before the debugger is even initialized
- acquire cmos spinlock when accessing the cmos registers
- Fix amd64 build
svn path=/trunk/; revision=53631
- Take NetworkAddr into account when a named pipe client is opened.
- Return RPC_S_SERVER_UNAVAILABLE when an attempt to create a named pipe client fails with an ERROR_BAD_NETPATH error.
This fixes the first test failure in the advapi32 service winetest.
svn path=/trunk/; revision=53630
- More uninitialized value fixing. Again Amine Khaldi spotting them with his magical powers.
- I took the liberty to convert tabs I found to spaces. Normally I shouldn't mix formatting with coding changes but I forgot. Sorry.
svn path=/branches/shell32_new-bringup/; revision=53629
- Attempt at fixing bug 6315 (it doesn't completely work yet). Original patch by Edijs Kolesnikovics.
svn path=/branches/shell32_new-bringup/; revision=53628
- give apic hal its own halinit_apic
- give minihal an own halinit_mini.c containing function stubs
- move HalInitializeProcessor to halinit.c and add hal specific function HalpInitProcessor
- get rid of all the #ifndef _MINIHAL_ in halinit.c and move the file to lib_hal_generic
- improve the hal version/build check
- Update spec file for amd64
svn path=/trunk/; revision=53625
- Add a test case for SetActiveWindow that shows its behavior related to owner windows. It should point the bug observed in bug 1239
svn path=/trunk/; revision=53621
Move the internal headers below the redefinition of types, because gcc instantiates inline function where they are defined, using the original types. Fixes build.
svn path=/trunk/; revision=53620
Make KeNumberProcessors internally an UCHAR and move it into krnlinit.c instead of having it once per architecture. Fixes a bunch of MSVC warnings.
svn path=/trunk/; revision=53618
- No code changes. Make the code more readable in preparation for investigating some failing wine tests.
svn path=/branches/shell32_new-bringup/; revision=53617
- move v86 code out of trap.S into v86.S (APIC uses a different trap entry file), compile it directly in the dll instead of a lib to be able to silence linker warnings
- Group files in HAL_PIC_SOURCE and HAL_APIC_SOURCE, you just need to replace PIC with APIC to build the hal with APIC support (we can later use it to build the other hals: halapic, halaacpi, halmacpi, halmps)
- use apic on amd64 builds
- give halacpi its own resource file
- silence MSVC linker warnings
svn path=/trunk/; revision=53616
- Fix max key name length, max value name length, max value data length properties of registry key not being set correctly
- Fixes#6042 - regedit on LiveCD shows keys and values now
svn path=/trunk/; revision=53614
Start implementing APIC support, which is needed for both SMP and x64. It will use the local APIC + I/O APIC for interrupt control, the RTC instead of the PIT for the timer interrupt (PIT doesn't always work with I/O APIC), the APIC timer for profiling and finally the TSC for the performance counter and KeStallExecutionProcessor.
The code is incomplete and doesn't work yet
svn path=/trunk/; revision=53611
- Add missing initialization. Fixes the problem where explorer fails to show the child window at the first attempt to open a folder
svn path=/branches/shell32_new-bringup/; revision=53610
- Any attempt to open or create a file on a remote machine will return STATUS_BAD_NETWORK_PATH. This means the server is not available. Trying to create \\testserver\testshare\testdir\test.txt fails just like on Windows XP.
- Add service key and values for the MUP driver. Add the driver to the File System group because adding it to the Network group, like on Windows XP, does not load the driver.
svn path=/trunk/; revision=53609
- Use functions instead of macros for irq<->vector<->irql translation, this will be neccessary, once apic support is there
- Disable mp hal in rbuild for now, its broken anyway
svn path=/trunk/; revision=53600
- Remove unnecessary assignments. Spotted by PVS and Dmitry Chapyshev. This may change the behaviour of that codepath, so test results are going to be observed.
svn path=/trunk/; revision=53596
- Fixed over check conditions for deleting regions in set window region.
- Disabled offsets with window region. Fixes wine tests..
svn path=/trunk/; revision=53595
RStartServiceA: Convert the ANSI start arguments vector to a UNICODE vector and pass it to ScmStartService. Calls to StartServiceA can now be used to pass arguments to a starting service.
svn path=/trunk/; revision=53579
- add flags to ASM as well instead of hacking the compile rule
- remove -fPIC flag from ASM compile flags, (it's a cmake bug)
- enable languages before including compiler specific files.
svn path=/trunk/; revision=53567
- Add ObReference to build. Was missing for some reason
- Use the new CMake macros in preparation for PCH
- Use kmtest_.exe as a file name for testcd to please rosautotest. Perhaps there's a better solution later
svn path=/branches/GSoC_2011/KMTestSuite/; revision=53562
- Replace add_linkerflag and add_compiler_flags with their new equivalents add_target_link_flags/add_compile_flags
- Remove those now obsolete macros
- Use the new set_module_type to get rid of evil hacks for drivers without entry point (wmilib, drmk)
svn path=/trunk/; revision=53558
- merge set_module_type function into the global CMakeMacros.cmake file, leaving compiler particular bits to set_module_type_toolchain function.
- Add the ENTRYPOINT and BASEADDRESS options to this function. Not used yet
svn path=/trunk/; revision=53553
- merge r53215 and r53221 from themes branch
- Fixes drawing apps that use SetWindowRgn and also fixes a gdi object leak
See issue #1239 for more details.
svn path=/trunk/; revision=53548
- Add add_target_property macro and add wrapper macros to easily set important properties
- This makes macros more consistent, follows CMake property names, favors per-target instead of per-directory property settings and favors adding to properties instead of replacing them
- Convert the global settings to use the new add_compile_flags and add_target_link_flags - no functional change
- Fix add_importlibs to add to compile definitions instead of replacing them
- Big thanks to Amine
svn path=/trunk/; revision=53545
- base/application UTF-8 conversion patch by Elton Chung;
- minor corrections by me;
See issue #6333 for more details.
svn path=/trunk/; revision=53544
- Fix bug related to correct error code returning in delete_files. The value of 1026 was revealed to be returned by windows 2003 server. Score several passed winetests.
- Fix a bug in ShellLink::SetShowCmd and score one more passed winetest
svn path=/branches/shell32_new-bringup/; revision=53543
- Introduce WINE-like debug macros ERR, FIXME, WARN and TRACE
- Always show ERR and FIXME information
- WARN and TRACE are enabled by DebugPrintMask in debug.c as before
- Thanks to new macros find bug in CMake livecd creation and fix it
svn path=/trunk/; revision=53542
- Don't do some parameter checks, even if MSDN says they're incompatible. Winetests with windows 2003 server as reference tells us msdn lies. (Spotted by Victor Martinez)
svn path=/branches/shell32_new-bringup/; revision=53541
- Romanian syssetup and usetup translation by Stefan Fulea <fulea.stefan> at <gmail.com>;
- Fix ipconfig build under MSVC, use "" instead of \" as escape character;
svn path=/trunk/; revision=53539
[FORMATTING]
- Format the code to a more acceptable level. This is just for my sanity while sifting through it.
svn path=/branches/shell32_new-bringup/; revision=53537
- rewrite the parser to handle command line argument parsing
- include basic description of what the rules for parsing are
- some minor formatting here and there (couldn't be helped)
svn path=/branches/shell32_new-bringup/; revision=53533
- Move cleaning the shared user data to after WinLdrSetProcessorContext, as we need the new page tables to be active
- Fix indentation/coding style in registry.c
- Fix a few 64 bit warnings
- delete <arch>/ntsetup.c and rename <arch>/wlmemory.c to <arch>/winldr.c
svn path=/trunk/; revision=53526
Don't use enums as bitfields. MSVC doesn't handle it the way gcc does. Fixes clock initialization on MSVC builds. You would wonder how the MSVC build even worked without a clock interrupt :D
svn path=/trunk/; revision=53524
- Massively refactor the winldr code
- move common parts of LoadReactOSSetup and LoadAndBootWindows into LoadAndBootWindowsCommon
- Combine architecture specific code into 2 functions: WinLdrSetupMachineDependent prepares the main stuff and WinLdrSetProcessorContext is the last thing done before transferring control to the kernel.
- rename WinLdrTunOnPaging to WinLdrSetupMemoryLayout
- Stop wasting stack space by decreasing the number of huge text buffers used
- Don't handle x86 specific data like Tss in portable code
- Add the progressbar for reactos setup as well
- Add missing DPRINT_PELOADER to DEBUG_ALL
svn path=/trunk/; revision=53510
This will enforce use of reactos msvcrt when apps provide one.
Its required because cmake dlls currently depend on msvcrt for their startup code.
svn path=/trunk/; revision=53505
- use /h to specify header file name when producing rpc proxy file.
Don't ask me why /header doesn't work, I have no idea.
svn path=/trunk/; revision=53504
Until now, rpcrt4 winetests where badly broken : client calls remotely forwarded to serverside functions which called... client side functions. One has to wonder how testbot survived the operation
- Cmake : fix that
- rbuild : add a hack keeping the old behaviour, because I don't know how to make this right
svn path=/trunk/; revision=53503
- Add stub hidusb driver
- Hidusb driver is reponsible to connect with the hidclass driver (not yet present) and connect hid framework with usb hid devices
svn path=/branches/usb-bringup/; revision=53500
We cannot make any assumptions about the latency whith which the timer interrupt fires after a rollover, since VBox (other VMs probably as well) doesn't always meet this. Add another check to KeQueryPerformanceCounter that gracefully handles missing interrupts. Also raise to DISPATCH_LEVEL, since the function is not reentrant.
svn path=/trunk/; revision=53496
- revert r53355.
- it's changes are not needed as wine's spy channel is better, and this time we will get rid of all unwanted changes ( my previous commit did not fix all unwanted changes)
svn path=/trunk/; revision=53492
* Revert r53487 (and as a result 53490) as per Giannis (the committer) suggestion. It broke testbot and he couldn't look into it at the moment.
svn path=/trunk/; revision=53491
- reenable -Oif flag, it's too good to be disabled
With this commit, advapi32:service winetest suffers only 2 failures.
svn path=/trunk/; revision=53481
- duplicate type entry for parameters with the range attribute.
This fixes the problem that Eric Kohl detected.
This has already been sent to wine.
svn path=/trunk/; revision=53480
- Report a unique instance ID that can be used to identify different processors
- Device Manager now displays one processor device per CPU
svn path=/trunk/; revision=53478
Revert parts of r53171:
Remove the -Oif flag for WIDL because the NT4 stub code generated by the current WIDL is heavily broken with respect to the range attribute and others. Please enable the -Oif feature only after these bugs are fixed in WIDL!
svn path=/trunk/; revision=53469
- Clean up win32k headers even more
- remove INTERNAL_CALL and use NTAPI instead
- remove several unneeded definitions
- remove several unneeded header inclusions
svn path=/trunk/; revision=53467
- fix a bug in x86 PnpBiosGetDeviceNodeCount
- enable pxe for amd64 (not yet implemented though)
- store boot drive and boot partition in shared memory isnetad of relying on a register, which seems to be erased on switch to long mode
64 bit freeldr is back on track!
svn path=/trunk/; revision=53459
- Remove /nologo option from RC again - WDK's version doesn't support it
- Add /nologo to MIDL's flags
- Remove redundant IDL compiler declaration for GCC build
svn path=/trunk/; revision=53447
- Fix wrong loop condition which would often lead to heap underread.
- Fix wrong subkey string length calculation, which would result in an incorrect string being used to open an image specific key.
- Don't close the key handle in LdrQueryImageFileKeyOption() because it's used more than once and closed after it's not needed anymore.
- Use Zw* functions instead of Nt* where necessary in LdrQueryImageFileKeyOption().
- Per-image execution options work now (e.g. enabling DPH).
svn path=/trunk/; revision=53446
- Implement real mode entry point for amd64 (amd64 freeldr starts but dies later, it cannot yet switch back to real mode, which is a bit more complicated)
- delete empty folders
svn path=/trunk/; revision=53443
- Resize controls so text fits in
- Improve Polish translations
[MSPAINT]
- Use GET_X/Y_LPARAM macros instead of LOWORD/HIWORD in mouse messages because coordinates can be negative
svn path=/trunk/; revision=53441
- According to MSDN IOCTL_VIDEO_QUERY_POINTER_CAPABILITIES doesn't have an in-parameter. This code was copied from MS sample code, but it doesn't make sense, since the size field doesn't even match the in-buffer size. VBox driver also doesn't use these parameters and we don't handle the ioctl anyway.
svn path=/trunk/; revision=53434
- Use sizeof(PVOID) instead of sizeof(PFOO) just for clarity that we really want a pointer size
[SCSIPORT]
- Don't initialize a pointer with RtlZeromemory
svn path=/trunk/; revision=53429
Add PCI_HAL device for non-ACPI machines. Code taken from the ACPI-HAL. This is just a quick hack and needs some more improvements.
svn path=/trunk/; revision=53425
- Handle IRP_MN_QUERY_CAPABILITIES for the ACPI_HAL device object
[PNPMGR]
- Don't bring up a device if a required IRP has failed
- Don't perform operations on a child if it is not ready yet
svn path=/trunk/; revision=53421
- Do not return an NTSTATUS from KeSynchronizeExecution, which returns BOOLEAN
[KERNEL32]
- Do not return an NTSTATUS from BasepGetModuleHandleExW
Found by MSVC.
svn path=/trunk/; revision=53413
- fix usage of set_entrypoint
- link libpng to ntdll on amd64
- build crtdll and nmidebug only on x86
- silence some linker warnings on MSVC
- compile riched20/msvc-thiscall.c only on x86
- Add WIN64 definition to acpica
- Add ehandler.c to libcntpr
svn path=/trunk/; revision=53396
- Fix copy paste error in DIALOG_IsAccelerator, fixes a crash.
- Up Dialog.c to 1.3.26, sync port and minimized differences.
svn path=/trunk/; revision=53391
- Implement FlashWindowEx without the state verification test via global atom and window properties list. Pass all but two in win:test_FlashWindowEx.
- Add missing atom's and global thread information pointer, based on patch from bug 5655.
svn path=/trunk/; revision=53385
- Up win to 1.3.26, keeping original patch plus skipping test_SetForegroundWindow, it hangs on WaitForSingleObject line 2771, needs regression testing.
svn path=/trunk/; revision=53384
- Print an debug message if IRP_MN_FILTER_RESOURCE_REQUIREMENTS fails
- Don't let a broken driver remove our resource requirements list
svn path=/trunk/; revision=53379
* CreateFile: If the specified file or device does not exist, the function fails and the last-error code is set to ERROR_FILE_NOT_FOUND. Spotted by Claudiu and Vic.
svn path=/trunk/; revision=53373
- Mark ExTimer test as hidden
- Mark some more tests as interactive in ExHardError
- Now an automated test run on ROS runs smoothly, aside from assertions
svn path=/branches/GSoC_2011/KMTestSuite/; revision=53365
- Add current source directory to MS RC include paths
- Add include directories to WRC include paths
- Now they should behave identically enough. Thanks to Amine.
svn path=/trunk/; revision=53361
Rewrite the command packet transfer code between the service control manager (services.exe) and the service control dispatcher (advapi32.dll). The service name is no longer stored in an array of WCHARs at the end of the command packet. The service name is now stored behind the fixed size command packet header using an offset value in the header to determine the start of the string. The same method will be used to store start arguments to the service dispatcher, but it is not fully implemented yet.
svn path=/trunk/; revision=53358
QueryServiceLockStatusA/W: If lpLockStatus is NULL or cbBufSize is less than sizeof(QUERY_SERVICE_LOCK_STATUSA/W) pass a pointer to an buffer to RQueryServiceLockStatusA/W.
svn path=/trunk/; revision=53356
- Remove some unneeded and misleading debug prints when parsing debug channels
- Switch NtUser to the new debug system
svn path=/trunk/; revision=53354
- Don't mess with the device interface string
- Fixes receiving power button events
- The "ACPI Shutdown" option in the VirtualBox menu works now
svn path=/trunk/; revision=53351
Fix completely non-standard, broken and retarded definition of GET_UNALIGNED_W for 64 bit architectures
The function is supposed to read an unaligned little endian USHORT value. The old version copied the 2 bytes with memcpy to an aligned stack variable and then (bug bug) read the unaligned value directly (discarding the result of the previous operation) and passed it to bswap if neccessary.
The new function simply reads the 2 bytes and combines them using << and |.
svn path=/trunk/; revision=53348
- Add __C_specific_handler stub
- Add aliases for C++ code on amd64 builds
- Fix fmod symbol
- Merge .CRT section into .data on amd64
- remove an old hack
svn path=/trunk/; revision=53342
- Fix acquiring image path from process parameters block (the if condition was inverted so that normalization didn't occur when it should).
- Reenable the image options initialization code.
svn path=/trunk/; revision=53327
- Rename memgen.c to byteswap.c and merge with the implementations from largeint.c, that were using intrinsics
- Fix amd64 build
svn path=/trunk/; revision=53321
- Implement a special case copy-in for IRP_MN_FILTER_RESOURCE_REQUIREMENTS to prevent NULL pointer access in function drivers
svn path=/trunk/; revision=53304
- Fix a typo in a remove lock timeout check
- Fixes breaking into kdbg randomly when a caller specifies no timeout on their remove lock
svn path=/trunk/; revision=53302
- Make IRP major function name table public
- IoHelper: catch all IRPs and add debug messages
svn path=/branches/GSoC_2011/KMTestSuite/; revision=53293
- ObType: take advantage of r53211
- IoDriverObject: add the remaining functions from the old drvobj and devobj tests
- fix some rbuild files
svn path=/branches/GSoC_2011/KMTestSuite/; revision=53280
- Add a dummy IoHelper driver that can be loaded by driver/device tests
- Minor changes to other standalone drivers
svn path=/branches/GSoC_2011/KMTestSuite/; revision=53177
- More reworking of ObType test. Now it works reliably on Windows.
- Split out testing of object referencing into separate ObReference test
svn path=/branches/GSoC_2011/KMTestSuite/; revision=53169
- Remove breakpoint on buffer overflow
- Use a compromise for ok_* macros for SIZE_T and [U]LONG_PTR values
svn path=/branches/GSoC_2011/KMTestSuite/; revision=53167
- Rename ObCreate test to ObType, and apply some major refactoring. Now _almost_ works on Windows....
- Add a readme file
svn path=/branches/GSoC_2011/KMTestSuite/; revision=53103
- add IoDriverObject test, partly ported from old kmtests. Should possibly be merged with another test
- ExDoubleList: add another small testcase
svn path=/branches/GSoC_2011/KMTestSuite/; revision=53047
- add RtlSplayTree test. Patch by arty (hacks by me ;])
- add RtlAvlTree test using the same code
- add kmtest_drivers and kmtest_all build targets
svn path=/branches/GSoC_2011/KMTestSuite/; revision=53046
- add kmt_platform.h that includes user or kernel headers as appropriate and allows Rtl tests to run in user mode without modification
- include kmt_platform.h from kmt_test.h, so that tests don't have to include separate headers. This also allows for a PCH
svn path=/branches/GSoC_2011/KMTestSuite/; revision=53021
- probe user address in user mode
- don't free the result buffer before the driver unlocks it. Fixes assertion in ROS for large buffers
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52947
- add KeSpinLock test
- KeIrql: also test the actual stdcall versions of KeRaiseIrql/KeLowerIrql
- ExHardError: take advantage of r52847
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52869
- KeIrql: make use of the new build type features
- KeApc: test internals of enabling/disable APCs
[KMTESTS/EX]
- better portability for ExInterlocked test
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52782
- small structure improvement in kmt_test.h
- add globals to allow easy checking for UP/MP & Free/Checked builds, as that is required for some tests
- add KernelType example test demonstrating the above
- add a function to restore the irql (allows some failing tests to exit gracefully)
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52781
- put driver files into 'bin' with the other tests instead of system32\drivers
- globally define NTDDI_VERSION=NTDDI_WS03_SP1 instead of doing it in each test
- fix testlist sorting (ASCII-order, but ignore leading minus)
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52780
- Add some enums which were accidentally removed.
- Add the rather strange Wineisms which some of their dlls rely on.
- Should fix the build for this branch
svn path=/branches/shell32_new-bringup/; revision=52750
- ExResource part 2: test concurrent acquisition of a resource in multiple threads
- ExInterlocked part 2: check calling convention correctness
- these show quite a number of bugs in ReactOS ;)
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52674
- decrease log buffer size because ROS has a problem with unlocking it otherwise ;)
- add test for critical/guarded regions
- add test for executive resources (part 1)
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52592
- prepare the device object for being opened multiple times (required for communication with special-purpose drivers)
- more verbose DPRINTs
- implement utility functions for app<->driver communication
- misc fixes
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52541
- rework the kmtest loader application to follow everything else's coding style and to allow user-mode test parts
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52519
- add skip() functionality (with more useful syntax than Wine's)
- add format attributes for GCC
- do not show the full path of the source file for failed tests
- general cleanup
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52489
- add START_TEST macro in case test functions ever have to be passed an argument
- add a few ok_* convenience macros
- remove unnecessary C library function renaming macros
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52371
- add the proper testing framework functions for winetest-like feel
- remove thereby obsolete Log* functions
- update example test to show usage of the new macros
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52359
- reference streamout.c and provide a wctomb stub to have an any-IRQL vsnprintf implementation for the test framework
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52330
- use a shared memory buffer for storing test results to provide seamless communication between all test parts
- Now user mode code can easily add messages to the buffer
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52216
- add a minimal example test so there's something to see when running kmtest
- remove _DLL and __USE_CRTIMP in accordance with r52213
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52215
- add simple loader application that can list/run tests from a driver via DeviceIoControl, and prints output to the console
svn path=/branches/GSoC_2011/KMTestSuite/; revision=52157
- Don't leave DISPATCH_LEVEL while holding a spin lock acquired at DISPATCH_LEVEL
- Synchronize cancellation checking by acquiring the cancel spin lock
svn path=/branches/usb-bringup/; revision=52071
- Comment out getting usb device interface. It is currently unused and usbhub does not yet support it
- Forward pnp capabilities request to lower device object
- Fix compilation by Usurp
svn path=/branches/usb-bringup/; revision=52063
- Copy disk.inf from trunk
- Add upper layer filter partmgr when usb mass storage device is plugged in
- Usbstor now loads and starts up
- Currently stops at loading partmgr(not yet present)
svn path=/branches/usb-bringup/; revision=52039
- Check if the port is usb 2.0 controller. In that case call the Usb2.0 initialize function
- Prevents crash when used with usbohci
svn path=/branches/usb-bringup/; revision=52036
- Fix OHCI_ISO_TD structure. Fixes host system crashes in Windows XP
- Fix multiple bugs in isochronous transfer implementation
- Still not yet working, as the interrupt completion is not fired yet
svn path=/branches/usb-bringup/; revision=51998
- Implement function to retrieve isochronous head endpoint descriptor
- Implement function to retrieve current frame number
- Set isochronous descriptor flag for head endpoint
- Implement retrieve device status for devices
- Implement retrieve class status from device
- Partly implement isochronous descriptor handling in usb queue
- Start implementing isochronous transfers in ISUBRequest
- Code currently not tested as the Virtual Machine with XP + ReactOS usbohci driver brings down the host system when starting the iso transfers. Ironically it crashes in MS usbohci driver
svn path=/branches/usb-bringup/; revision=51957
- Merge 51783
- The final revision to merge from the existing shell32 code.
- The C++ shell32 module is now fully up to date and ready for testing.
svn path=/branches/shell32_new-bringup/; revision=51941
- Merge 51436, 51448 and 51526
- Properly support WM_SYSCOLORCHANGE in ShellView window. Changing system colors in ReactOS works perfectly now :)
svn path=/branches/shell32_new-bringup/; revision=51937
- Merge 51423, 51426 and 51435
- Remove "Play" and "Preview" translations - they shouldn't be in shell32
- Pass an argument array to FormatMessage to insert the filename details into error messages (instead of failing)
svn path=/branches/shell32_new-bringup/; revision=51934
- Set up periodic threshold (90 %)
- Add function to retrieve interrupt endpoints
- Add function to retrieve specific descriptors from interface. These are used by HID devices such as mice / keyboards
- Add function to retrieve interrupt interval
- Enqueue all endpoint descriptors at the end of the associated queue
- Only notify hardware of insertion when it is an bulk / control request
- Scan interrupt endpoint list to find the completed transfer descriptor
- Add debugging function to print out linked endpoint descriptors
- Interrupt transfers are now implemented.
- Tested in Windows XP SP3 + Vbox 4.04 + Microsoft 5-Button Mouse. The HID mouse installs, initializes and starts up. Unfortunately the mouse does not work as expected yet
svn path=/branches/usb-bringup/; revision=51922
- Remove dead code
- Silence traces
- Pass status & DoneHead as parameters to dpc routine
- Move IUSBRequest cleanup code into new function and use it for bulk / control transfer cleanup
- Fix bugs in AllocateEndpointDescriptor. It did not take the device address into account. It also did not respect the direction of the descriptor
- Implement support for bulk transfer requests
- Handle irp completion in CompletionCallback
- Tested in Windows XP SP3 + Vbox 4.04 + USB2.0 disabled + ReactOS usbstor + USB mass storage device
- OHCI Mass storage support is now also ready
- Next interrupt transfers
svn path=/branches/usb-bringup/; revision=51917
- Merge 51322 and 51323
- Return 1 in WM_ERASEBKGND when desktop is painted
- Fix background for desktop icons and their labels
- Moreover fixes explorer_new background
- Properly arrange icons in explorer shell view window. This also fixes navigating in files with keyboard arrows
svn path=/branches/shell32_new-bringup/; revision=51915
- Merge 50899, 50941, 50957, 50964, 51053
- Check the correct pidl instead of the free one
- Found by the amazing DPH!
- Return the icon index, not the icon id
-Fix missing "Open" menu entry on right click for All Users desktop shortcuts.
svn path=/branches/shell32_new-bringup/; revision=51910
- Merge 50114 and 50144
- Add NULL checks in IShellView and IDropTarget implementations. Should fix bug #5780.
- Bug 5756: TRANSLATION: multiple modules: Bulgarian by CCTAHEB
svn path=/branches/shell32_new-bringup/; revision=51909
- Merge 49969, refactored to use CComPtr's thanks to the new C++ model (it's sooo much cleaner now)
- Implement copy / paste for files when tree view is present
- Don't add shortcuts / executables to recent file list
svn path=/branches/shell32_new-bringup/; revision=51908
- Add glue code for supporting iso transfers
- Remove dead from Handle
- Remove broken asserts
- Add support for string descriptors
- Add support for class specific endpoint requests (Needs be ported to usbehci)
- Link to usbd driver
- Add support for retrieving string descriptor
- Rewrite configuration descriptor handling in IUSBDevice. New code is smaller, smarter and handles a lot more cases. Needs to be ported to usbehci
- Wrap usbdlib.h include in extern c macro, it has c linkage
- Control transfers should now work
- Need to implement support for isochronous / bulk / interrupt transfers until functional
- Tested with Bluetooth USB Stick (multi function, interrupt / bulk / control / isochronous) / USB Microphone (isochronous & control) in XP SP3
svn path=/branches/usb-bringup/; revision=51900
- Add many missing (or previously hidden) shell COM interfaces to the public headers.
- The new C++ shell32 dll should now build.
svn path=/branches/shell32_new-bringup/; revision=51896
- Add sanity checks
- Preserve command status value when notifying the hc that a control / bulk endpoint was added
- Re-enable root hub notification interrupt when reset port has been completed
- Dispatch processing to USBQeueu when DoneHead event arrives
- Scan endpoints to find the logical endpoint which was completed by the hardware
- Signal completion to IUSBRequest by calling CompletionCallback
- Create a final transfer descriptor which is linked to the endpoint descriptor
- Control transfers now appear to be working (Device retrieves device descriptor / configuration descriptor and succeeds in setting device address)
- Configuration parsing code needs more work now (currently fails there due to invalid / unexpected configuration descriptor from device)
svn path=/branches/usb-bringup/; revision=51888
- Enable global power mode
- Wait untill reset is complete
- Clear reset complete bit when reset is done
- Enable port
- Reset port now works
- USBOHCI still hangs after adding first control request, needs more investigation
svn path=/branches/usb-bringup/; revision=51883
- Start implementing control transfers
- Move initialization of usb queue after the bulk / control head endpoints have been created
- Add interface functions to retrieve bulk / control head endpoint descriptors
- Add macros for setting transfer descriptors field (taken from Haiku)
- Partly implement IUSBQueue::AddUSBRequest for control / bulk transfers
- Create endpoint descriptor for control request and link setup descriptor to it
- Link setup descriptors to data descriptor / status descriptor
- WIP (needs fixes)
svn path=/branches/usb-bringup/; revision=51864
- Fix bug in calculating offset of status register
- Start implementing deferred processing routine
- Signal status change when new device is arrived
- Implement basic GetPortStatus
- Start implementing SetPortFeature, ClearPortStatus
- ResetPort not yet fully working, WIP
svn path=/branches/usb-bringup/; revision=51858
- Add check that controller is functional state
- enable interrupts after the controller has been started
- partly implement isr routine
svn path=/branches/usb-bringup/; revision=51852
- Start of OHCI support
- Based on new usbehci driver and Haiku USB Stack
- Driver initializes the controller and retrieve number of ports
svn path=/branches/usb-bringup/; revision=51851
- Pass memory manager to USBQueue object
- Fix bug in memory manager which did not check if an allocation equals page size
- Implement interrupt queue head array with frequencys of 1ms up to 32ms
- Store queue heads in the sync schedule array
- WIP
svn path=/branches/usb-bringup/; revision=51826
- Add error handling to USBSTOR_CSWCompletionRoutine for errors that can be recovered via reset of pipe/port/controller. Only pipe reset is attempted. WIP.
svn path=/branches/usb-bringup/; revision=51808
- Implement retrieving serial number from device
- Use serial number to format instance id and retrieve in StorageDeviceProperty requests
- Silence traces
svn path=/branches/usb-bringup/; revision=51770
- Add asserts
- Rewrite queue methods to accept the FDO only
- Fix multiple synchronization bugs
- Don't start next irp before current irp has completed
- Use contiouslogicalbytes blocks from srb
- Should fix race conditions
svn path=/branches/usb-bringup/; revision=51764
- Silence warnings
- Silence traces
- Add asserts
- Add code to detect usb reset in progress. In that case cancel the request
- Add more code to synchronize urb action
- WIP, still hangs are observed
svn path=/branches/usb-bringup/; revision=51759
- Start implementing error handling.
- On errors use a worker thread to attempt a recovery. If successful restart the operation that failed.
- If STALL_PID is received reset the pipe of the device.
- Not entered into code path yet. WIP.
svn path=/branches/usb-bringup/; revision=51754
- USBSTOR_SendRequest: The buffer for read/write may not be NonPagedPool, which is documented as a requirement for using MmBuildMdlForNonPagedPool. Also locking the buffers pages is also not an option as the routine is called at DISPATCH_LEVEL.
It so happens that Irp->MdlAddress is valid for read/write operations. Use it instead of procedure above.
- Add a sanity check to make sure the Mdl does describe the transfer buffer.
Fixes IRQL_NOT_LESS_OR_EQUAL bugcheck in windows. These changes also may fixed writing to device.
svn path=/branches/usb-bringup/; revision=51752
- Fix bug in read command, which uses wrong format. LBA address is stored in the CDB10 struct
- Don't build the mdl for nonpaged pool if the driver passes an mdl (i/o paging requests)
- Usbstor can now list disk contents and browse files when used in Windows XP
- Need to implement SCSIOP_WRITE to write files
svn path=/branches/usb-bringup/; revision=51694
- Handle SCSIOP_MEDIUM_REMOVAL
- Handle paging i/o requests. These request don't provide a data buffer in the srb, but the buffer is stored in the irp' mdl
- Fix the hack for unimplemented Mode Sense command
- Usbstor now completes initializes and receives read requests
- Still need to fix read request, as Windows XP fails to recognize the disk format, WIP
svn path=/branches/usb-bringup/; revision=51693
- Add a check at USBSTOR_FdoHandleStartDevice to make sure the device uses Bulk Transfers.
- Use macros for initializing the Urbs. The macros ensure that the proper fields of the URB are set correctly.
Fixes failing of getting device descriptor and getting devices BlockLength.
svn path=/branches/usb-bringup/; revision=51691
- Only get the string descriptors if they are available. Fixes crash when using usb mouse and keyboards in windows.
svn path=/branches/usb-bringup/; revision=51688
- Modify BuildBulkTransferQueueHead to support TransferBufferLengths larger than PAGE_SIZE * 5.
- Acquire a SpinLock before adding QueueHeads to AsyncList and PendingList.
- Dont request a new QueueHead for incomplete transfers in QueueHeadCompletion, as the memory for the just completed QueueHead has not been released yet. Doing so overwrites the m_TransferDescriptor[x] members with new address resulting in memory leaks. Instead request a new QueueHead after the QueueHead has been freed in QueueHeadCleanup.
- Fix a bug where a QueueHead was removed from the m_CompletedRequestAsyncList instead of the m_PendingRequestAsyncList.
- Temporary hackfix InternalCalculateTransferLength to return the TransferBufferLength. This hack will be removed as soon as possible.
- With these changes the hub and ehci driver allow viewing content of and transfers to/from usb disks.
svn path=/branches/usb-bringup/; revision=51684
- Save PortNumber in child device extensions as its needed when PNP request device capabilities.
- Handle IOCTL_INTERNAL_USB_SUBMIT_URB request by creating a new Irp and passing the Urb down to the miniport.
- Now working with our ehci driver, we at least see a usb driver and can browse directory in Windows, though doing so results in assert as ehci doesnt support transfers over 5 Pages yet. Fix coming soon.
svn path=/branches/usb-bringup/; revision=51672
- When clearing feature port reset, remove the flag from Change and if the port is hardware enabled set the enable flag in Status.
- When checking port status, if Change has any flags set an SCE request most be completed. Verified while writing usbhub driver in xp.
- Probably fixes a case where an endless loop of completing SCE requests.
svn path=/branches/usb-bringup/; revision=51642
- Implement constructing CompatibleIds string using device descriptor. However, if the Class of the device descriptor is 0, then Class, SubClass and Prot which make up this string must come from the Interface.
- Change Ids and Descriptions to Unicode Strings.
- Move the creating of Id strings to one place, CreateDeviceIds, and called it at device creation.
svn path=/branches/usb-bringup/; revision=51641
- Some devices implement a default Language, LangId of 0. Some do not. Use LangId 0x0409 (English) until LangId is implemented.
- Remove an unused variable.
svn path=/branches/usb-bringup/; revision=51622
- The lower level driver completes the SCE IRP whenever the change state has been modified for a port. When handling the SCE IRP completion only check for flags in change state. The Change flags must be cleared else the hub driver will get non stop SCE IRP completions.
- Allocate IRP from Pool vice calling IoAllocateIrp. Dont free the IRP in the completion routine as the port driver, oddly, frees them.
- GetUsbDeviceDescriptor: Remove use of MDL and use only buffer instead to make our new usbehci happy.
- When calling the interface routines, correctly use the BusContext returned from the interface instead of the RootHubPdo. This worked on windows as it just happened they were the same.
- Implement RootHubInitCallbackFunction, which only job currently is to send the first SCE IRP.
- For Start Device for child device objects fake success for now. Will be implemented later.
- Implement returning IRP_MN_QUERY_IDs, IRP_MN_QUERY_DEVICE_TEXTs and IRP_MN_QUERY_BUS_INFORMATION.
- Add basic handling for IRP_MJ_POWER.
- Misc code changes.
svn path=/branches/usb-bringup/; revision=51620
- Implement retrieving drive capacity
- Use constant from scsi.h instead of defining own format
- Don't reset returned parameters length to zero, as ioctl handler might set it
svn path=/branches/usb-bringup/; revision=51608
- Make DeviceId generation to match ms usbstor
- Don't hardcode the device type in USBSTOR_PdoHandleQueryHardwareId
- Don't hardcode device type in USBSTOR_PdoHandleQueryCompatibleId
- Hackfix generation of instance id
- Add comments
- Cleanup code
- Usbstor now receives ioctl from disk.sys
svn path=/branches/usb-bringup/; revision=51582
- Add UsbDeviceHandle to child device extensions structure to be passed with URB requests.
- Reimplement the Worker Thread DeviceStatusChangeThread called on SCE completion. Each SetFeature and ClearFeature resend SCE request.
Only create the usb child device if the PortStatusChange member of the device extensions was modified.
- Implement CreateUsbChildDeviceObject used for creating a deviceobject for the USB device and initializing it.
- Implement GetUsbStringDescriptor for getting DeviceId, TextDescription and InstanceId from USB Device.
- Move the initial QueryStatusChangeEndpoint to after the hub device receives start pnp.
- USBHUB driver is now back to the same functionality of previous version. Showing the device wizard for connected devices.
svn path=/branches/usb-bringup/; revision=51581
- Rewrite hardware id / instance id matching
- Implement support for compatible id
- UsbStor still not attached to disk.sys, needs more research on how to get this done
svn path=/branches/usb-bringup/; revision=51580
- Change parameter name for DeviceObject in functions to clearify which device object is to received IRPs.
- Fix a typo in GetPortStatusAndChange that caused the function to return nothing for status and change.
- Call QueryStatusChangeEndpoint to send the first SCE request in Query Bus Releations.
- Complete implementation of StatusChangeEndpointCompletion. Queue a work item for DeviceStatusChangeThread, which will be used to initialize the usb device and create related device objects.
svn path=/branches/usb-bringup/; revision=51576
- Don't free device descriptor if it failed to acquire it
- Don't move out of usb configuration array. Found by special pool
- Fix bug when storing interface information which lead to a overwrite
- Implement retrieving generic type for device
- Fix bugs in QueryDeviceId
- Implement querying hardware ids
- Implement querying instance ids
- With these changes, usbstor now initializes and create pdo, which is also enumerated
- Needs more code / fixes so that disk.sys attaches to usbstor
- WIP
svn path=/branches/usb-bringup/; revision=51575
- Invoke status change callback at passive level
- Fill out usb descriptor struct instead passing everything in char array
- Use 1 byte interrupt endpoint packet size for hub for now
- Fill in interface information when SelectConfiguration request is passed. Fixes sce callback and pool corruptions
- Return device descriptor in USBHI_QueryDeviceInformation
- Tested in WinXP with special pool enabled & overrun support
svn path=/branches/usb-bringup/; revision=51572
- Start work on reimplementing UsbHub driver.
- Implement SubmitRequestToRootHub, QueryStatusChangeEndpoint, QueryInterface, GetPortStatusAndChange, SetPortFeature and ClearPortFeature.Handle QueryBusRelations.
- Partially implement PNP handling for FDO. When starting the device get all the information about the RootHub the driver will need and enable Power on all ports.
- Partially implement PNP handling for PDO, QueryId and QueryDeviceText.
- WIP
svn path=/branches/usb-bringup/; revision=51569
- Store number of bytes transferred in the transfer descriptors
- Perform queue head completion when the door bell ring has been acknowledged. Fixes race condition between multiple irps in the async list
- Fix calculation of transfer length when the request is an bulk in operation
- Use EndPointDescriptor member to access transfer type / pid direction
- Use MmGetSystemAddressForMdlSafe to retrieve system address for urb buffer
- Fix check if first transfer buffer finishes on first size if the size is of page_size
- With these changes and little luck and good weather, usb mass storage devices have been seen to work in Windows XP SP3
- Code inspired of mjmartin usbehci driver and Haiku's usb stack
svn path=/branches/usb-bringup/; revision=51506
- Port URB_FUNCTION_CLASS_INTERFACE from mjmartin usbehci driver
- Fix typo in interface declaration
- Disable assert for now
- Return irp status pending when irp is added to the queue
- Fix bug in InternalGetPidDirection
- More work on the bulk transfers, not yet working
svn path=/branches/usb-bringup/; revision=51501
- Fix a type causing CommitIrp to fail.
- Add a sanity check in StatusChangeEndpointCallBack to check that a pending SCE irp is queued.
- Start Implementing Bulk Transfers. Needs more work.
svn path=/branches/usb-bringup/; revision=51493
- Set the NakCountReload to 0 in the QueueHead that is set in the Async register, as this is a dead QueueHead marked as halted and does not need it.
- Set the Async register in StartController. Note: The Async and Periodic registers can only be modified when the controller is stopped.
- Driver now works in Qemu.
svn path=/branches/usb-bringup/; revision=51491
- hub_controller.cpp: Change RtlSetBit and RtlClearBit to RtlSetBtis and RtlClearBits for Win2k compatibility.
svn path=/branches/usb-bringup/; revision=51488
- Implement retrieving device / configuration descriptor for usb devices
- Based on mjmartin usbehci driver
- Mass storage device found wizard now pops up (fails to install - WIP)
svn path=/branches/usb-bringup/; revision=51478
- Move error checking if controller reported an error inside the door bell ring check block
- Extend interface of IUSBRequest to include DeviceAddress, which is needed to retrieve configuration descriptor of newly set address
- Allocate CtrlSetup from nonpaged pool
- Release setup packet on cleanup
- Devices now receive an device address
- Currently asserts while retrieving configuration descriptor as this code path is not yet existant
*** Assertion failed: Urb->UrbHeader.UsbdDeviceHandle == NULL
*** Source File: d:\usb-bringup\drivers\usb\usbehci_new\hub_controller.cpp, line 1118
svn path=/branches/usb-bringup/; revision=51477
- Change interface to return real async queue head
- Set the async queue head register after the controller has been started
- Enable async queue in StartController
- Port DumpDeviceDescriptor from mjmartin usbehci driver
- Remove pseudo queue head from usb queue, instead use the real async queue head exported from IUSBHardwareDevice
- Get physical address for transfer data in BuildControlTransferQueueHead
- Retrieving device descriptor now ~works, currently stops at setting device address (needs more work)
svn path=/branches/usb-bringup/; revision=51476
- Check if the current request type is supported (currently only control is supported, bulk will be implemented soon)
- Fix bug in IUSBRequest::CreateQueueHead
- Code runs now until first device descriptor get request (async / periodic queue not yet enabled)
- mjmartin usbehci status not yet reached
- Let the fun now begin ;)
svn path=/branches/usb-bringup/; revision=51474
- Cache port status & port change settings because it appears that some ehci controllers do not store the status reset register bit correctly
- This commit is a temporary solutions, needs to be reworked
- Return correct status from HandleClassOther
- Handle IOCTL_INTERNAL_USB_SUBMIT_IDLE_NOTIFICATION to return status success
svn path=/branches/usb-bringup/; revision=51473
- Return the physical address of the async queue head in CUSBHardwareDevice::GetAsyncListRegister
- Remove superflous entries from queue head structure, they are processed within IUSBRequest class
- Remove USBHI_GetPortHackFlags, this function has been deprecated
- Add interface functions for IUSBRequest / IUSBQueue
- Callback into IUSBQueue when a hardware interrupt arrives
- Implement callback function to check for completed queue heads / free completed queue head depending on the Async Advance interrupt bit
svn path=/branches/usb-bringup/; revision=51466
- Create member for storing the request object
- Add interface function which is invoked when the queue head should be freed
- Add support functions which deal with completed queue heads
svn path=/branches/usb-bringup/; revision=51440
hub_controller:
- Implement StatusChangeEndpointCallBack called by HardwareDevice class when port status has changed on controller.
- Move Status Change Endpoint query code to IQueryStatusChageEndpoint as it is also needed by StatusChangeEndpointCallBack.
usb_request:
- Implement InternalCreateUsbRequest.
usb_queue:
- Implement CreateUsbRequest and AddUsbRequest
hardware:
- Implement GetAsyncListRegister and GetPeriodicListRegister.
- Implement SetStatusChangeEndpointCallBack for setting callback and context. Call the callback when a port status changes.
- Initialize the UsbQueue after creating the AsyncQueueHead, as the UsbQueue calls will call back with GetAsyncListRegister.
svn path=/branches/usb-bringup/; revision=51407
- Fix warning because comparing signed vs unsigned
- Implement function for retrieving the device descriptor
- Implement function for retrieving a configuration descriptor (including contained interface descriptor and endpoint descriptor)
- Cleanup interface for IUSBRequest:
- When caller initializes IUSBRequest with an IRP, then the operation mode is asynchronously. Therefore when the request is completed, IUSBRequest::ShouldReleaseRequestAfterCompletion will return true, which makes IUSBQueue call Release on IUSBRequest
- When the caller initializes IUSBRequest with an setup packet, the operation mode is synchronously. After submitting the IUSBRequest to queue, the caller should call IUSBRequest::GetResultStatus. This function will then block untill the operation has been completed. However, the caller needs to call Release() as those requests are not cleaned up by the IUSBQueue
svn path=/branches/usb-bringup/; revision=51406
- Initialize the MemoryManager only after allocating a CommonBuffer to pass to it.
- Create a QueueHead to put into the AsyncList Register.
svn path=/branches/usb-bringup/; revision=51402
- hub_controller: Implement Status Change Endpoint for RootHub.
Handling the URB can return STATUS_PENDING, as in the new SCE code. Check for this before completing the Irp.
Uncomment calls to PortStatus, SetPortFeature, and ClearPortStatus now that they are implemented.
- For function receiving a port number check that its not larger than the actual number of ports on the controller.
svn path=/branches/usb-bringup/; revision=51401
- Move USB_REQUEST_SET_FEATURE / USB_REQUEST_GET_FEATURE flags to header file.
- Remove QueueHead and Descriptor creation as its in UsbRequest.
- Move DMACommon Buffer allocation back to UsbHardwareDevice class.
- Implement Port Handling functions GetPortStatus, ClearPortStatus and SetPortFeature.
- Implement GetUSBQueue for returning pointer to UsbQueue and GetDMA to return DMAMemoryManager.
- Fix a typo in one of the defines for Port Register Flags.
svn path=/branches/usb-bringup/; revision=51397
- Use PQUEUE_HEAD as the Head of pending list and Async (Active) list vice using PLIST_ENTRY.
- Implement linking/unlinking a QueueHead to these Head QueueHeads.
- Implement chain linking/unlinking of QueueHeads. Thr plan being to used these functions during DMA operations to pull a number of QueueHeads from the pending list to place into the Async list for execution.
- Will look for optimization in these functions later.
svn path=/branches/usb-bringup/; revision=51390
- Use LIST_ENTRY in QueueHeads and Descriptors vice Next and Previous Pointers.
- Add functions to interface for setting AsyncListRegister and PeriodicListRegister.
- USBHardwareDevice: Initialize USBQueue when handling PNPStart.
- USBQueue: Allocate Common Buffer and use it to Create and Initialize DmaMemoryManager object.
- USBQueue: Implement CreateQueueHead and CreateDescriptor.
svn path=/branches/usb-bringup/; revision=51388
- Update interfaces with new methods needed for UsbHardWare class.
- Remove GetDmaMemoryManager as the DmaMemory will only be needed by UsbQueue class.
- GetDeviceDetails: Only fill in a parameter if its not null.
- Start implementing UsbQueue class.
svn path=/branches/usb-bringup/; revision=51386
- Implement IRP_MN_QUERY_DEVICE_RELATIONS for hub controller object
- Implement stub handler for URB_FUNCTION_CLASS_OTHER, URB_FUNCTION_SELECT_CONFIGURATION
- Partly implement URB_FUNCTION_GET_STATUS_FROM_DEVICE
- Return success from unimplemented USBHI_Initialize20Hub routine
- based on mjmartin usbehci driver
- Tested in Windows XP SP2
svn path=/branches/usb-bringup/; revision=51385
- Setup a default device descriptor for the root hub and initialize it with vendor & product id
- Partly implement URB_FUNCTION_GET_DESCRIPTOR_FROM_DEVICE, URB_FUNCTION_CLASS_DEVICE
- based on mjmartin usbehci
svn path=/branches/usb-bringup/; revision=51384
- Fix bug in CDMAMemoryManager initialization, which calculated the bitmap length wrong
- Create interface IUSBDevice, which will be used to abstract connected usb devices
- Implement support functions for the device interface.
- Implement USBHI_CreateUsbDevice, USBHI_InitializeUsbDevice, USBHI_GetUsbDescriptors, USBHI_RemoveUsbDevice, USBHI_GetExtendedHubInformation, USBHI_RootHubInitNotification, USBHI_SetDeviceHandleData, USBDI_GetUSBDIVersion, USBDI_IsDeviceHighSpeed
- Partly implement USBHI_QueryDeviceInformation
- Based on mjmartin usbehci
svn path=/branches/usb-bringup/; revision=51372
- Implement IDMAMemoryManager interface, with IDMAMemoryManager::Allocate, IDMAMemoryManager::Free
- Fix interface of IDMAMemoryManager
- Will be used by IUSBHardwareDevice
svn path=/branches/usb-bringup/; revision=51369
- Add reference count when the hub controller is created. Necessary to prevent the controller from getting deleted when the hub driver performs interface reference removes
- Implement IRP_MN_QUERY_ID, based on mjmartin ehci driver
- Fix bug in IRP_MN_QUERY_BUS_INFORMATION, it is supposed to return a PNP_BUS_INFORMATION not a simple guid
- Implement IOCTL_INTERNAL_USB_GET_DEVICE_HANDLE, IOCTL_INTERNAL_USB_GET_ROOTHUB_PDO, IOCTL_INTERNAL_USB_GET_HUB_COUNT based on mjmartin ehci driver
- Add interfaces for communicating with the hub driver
- Dispatch internal device requests not wmi requests (EHCI_Dispatch)
svn path=/branches/usb-bringup/; revision=51368
- Change VendorId and DeviceId to the same size as PciConfig members, PUSHORT.
- Add support function GetBusInterface.
- Implement GetDeviceDetails and ResetPort.
svn path=/branches/usb-bringup/; revision=51367
- Add flags for Port Status Control Register and remove structiures for Port Status Control Register.
- Remove USB Status Register and use the flags instead.
- Remove functions that are not used and never will be.
- Implement InterruptServiceRoutine and Deffered Procedure Routine.
- Partially implement port status change to detect device insert and removal.
svn path=/branches/usb-bringup/; revision=51366
- Add flags and structures needed for communicating with controller and handling schedules.
- Add support routines for modifying operational registers on controller.
- Implement getting the controller capabilties, starting and stopping the controller.
svn path=/branches/usb-bringup/; revision=51365
- Create Interface IDispatchIrp which is used to handle ioctl requests
- Partly implement IHubController interface
svn path=/branches/usb-bringup/; revision=51355
- Define tag for allocations
- Fixes assert hit in ReactOS
- Fix typo in IHubController interface declaration
svn path=/branches/usb-bringup/; revision=51353
- Fix IOCTL_GET_HCD_DRIVERKEY_NAME implementation
- Now needs IOCTL_USB_GET_ROOT_HUB_NAME to get further in usbview
- Tested in Windows XP SP2
svn path=/branches/usb-bringup/; revision=51350
- Build fix#1 for mingw
- Define cxa_pure_virtual which is required that gcc is aware that it should really use the libc++ instead of falling back to libc
svn path=/branches/usb-bringup/; revision=51348
- Implement CHCDController::Initialize, CHCDController::HandlePnp for PDO / FDO,
- Partly implement CHCDController::HandleDeviceControl
- Based on mjmartin usbehci sources
- Driver now startups and acquires pnp resources
- Fails in HandleDeviceControl with IOCTL_GET_HCD_DRIVERKEY_NAME when usbview is launched
- Tested in Windows XP SP2 + Vmware Server 2.0.2
svn path=/branches/usb-bringup/; revision=51343
- Create empty USB EHCI driver in c++
- Design & review the interfaces with mjmartin
- Create stubs for creating IUSBHardwareDevice / IHCDController classes
- Implement DriverEntry, EHCI_AddDevice and EHCI_Dispatch
svn path=/branches/usb-bringup/; revision=51335
2011-04-14 11:35:16 +00:00
15773 changed files with 2216765 additions and 2878084 deletions
IDS_STRING_LICENSE "ReactOS kalkulaèka je svobodný software vydaný pod licencí GNU GPL.\r\n\r\nZde najdete znìní GNU GPL licence:\r\nhttp://www.gnu.org/licenses/gpl.html\r\n\r\nZde také najdete pøeklad GNU GPL licence:\r\nhttp://www.gnu.org/licenses/translations.html"
IDS_STRING_LICENSE "ReactOS kalkulačka je svobodný software vydaný pod licencí GNU GPL.\r\n\r\nZde najdete znění GNU GPL licence:\r\nhttp://www.gnu.org/licenses/gpl.html\r\n\r\nZde také najdete překlad GNU GPL licence:\r\nhttp://www.gnu.org/licenses/translations.html"
IDS_STRING_LICENSE "ReactOS Calc is free software released under GNU GPL license.\r\n\r\nYou can get a copy of GNU GPL license here:\r\nhttp://www.gnu.org/licenses/gpl.html\r\n\r\nYou can also get a translation of GNU GPL license here:\r\nhttp://www.gnu.org/licenses/translations.html"
MENUITEM "A ReactOS &Számológépről", IDM_HELP_ABOUT
END
END
// String Tables
STRINGTABLE DISCARDABLE
BEGIN
IDS_STRING_LICENSE "A ReactOS Számológép a GNU GPL licenc alatt kiadott szabad szoftver.\r\n\r\nA GNU GPL licenc egy másolatát innen szerezheted be:\r\nhttp://www.gnu.org/licenses/gpl.html\r\n\r\nValamint a GNU GPL licenc fordítását innen szerezheted be:\r\nhttp://www.gnu.org/licenses/translations.html"
IDS_STRING_LICENSE "Calculadora ReactOS é um software livre lançado sobre a licença GNU GPL.\r\nVocê pode adquirir uma cópia da licença GNU GPL no endereço abaixo:\r\nhttp://www.gnu.org/licenses/gpl.html\r\n\r\nVocê também pode adquirir uma tradução da licença GNU GPL no endereço abaixo:\r\nhttp://www.gnu.org/licenses/translations.html"
IDS_STRING_LICENSE "ReactOS Calc este software liber sublicența GNU GPL. \r\n\r\nGăsiți o copie a acestei licențe aici:\r\nhttp://www.gnu.org/licenses/gpl.html\r\n\r\nDe asemenea găsiți traduceri ale aceste licențe aici:\r\nhttp://www.gnu.org/licenses/translations.html\r\nTraducerea în limba română de Petru Dimitriu."
IDS_STRING_LICENSE "Această aplicație este publică; fiind permisă modificarea și/sau (re)distribuția sa în termenii Licenței Publice Generale GNU publicată de Free Software Foundation; sau versiunea 2 a Licenței, sau (la alegere) a oricărei versiuni ulterioare.\r\n\r\nAceastă aplicație este distribuită doar în speranța de a fi utilă, FĂRĂ însă NICI O GARANȚIE; nici măcar cu garanția implicită a VANDABILITĂȚII sau a UTILITĂȚII ÎNTR-UN SCOP ANUME. Pentru mai multe detalii consultați Licența Publică Generală GNU.\r\n\r\nPuteți vedea această licență aici:\r\nhttp://www.gnu.org/licenses/gpl.html\r\n\r\nDe asemenea puteți consulta traduceri neoficiale ale acestei licențe aici:\r\nhttp://www.gnu.org/licenses/translations.html"
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."
IDS_LICENSE "Ce programme est un logiciel libre ; vous pouvez le redistribuer et/ou le modifier tout en respectant les termes de la ""GNU General Public License"" publiée par la Free Software Foundation; dans sa version 2 (ou selon votre préférence) toute version ultérieure.\r\n\r\nCe programme est distribué dans l'espoir qu'il sera utile, cependant SANS GARANTIE D'AUCUNE SORTE ; sans même une garantie implicite de COMMERCIABILITÉ ou DE CONFORMITÉ À UNE UTILISATION PARTICULIÈRE. \r\n\r\nVoir la Licence Publique Générale GNU pour plus de détails. Vous devriez avoir reçu un exemplaire de la Licence Publique Générale GNU avec ce programme ; si ce n'est pas le cas, écrivez à la Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA."
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."
IDS_LICENSE, "Questo programma è software libero; può redistribuirlo e/o modificarlo sotto\ni termini della licenza pubblica GNU come pubblicata dalla Free Software Foundation; sia la versione 2 sia una versione successiva (a sua scelta).\r\n\r\nQuesto programma è distribuito\nnella speranza che sia utile, ma SENZA ALCUNA GARANZIA; senza neanche la garanzia implicita\ndi NEGOZIABILITA' o APPLICABILITA' per un particolare scopo. Si veda la licenza generale pubblica GNU per maggiori dettagli.\r\n\r\nDovrebbe aver ricevuto una copia assieme a questo programma; se così non fosse, scriva alla Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA."
IDS_LICENSE "Questo programma è software libero; può redistribuirlo e/o modificarlo sotto\ni termini della licenza pubblica GNU come pubblicata dalla Free Software Foundation; sia la versione 2 sia una versione successiva (a sua scelta).\r\n\r\nQuesto programma è distribuito\nnella speranza che sia utile, ma SENZA ALCUNA GARANZIA; senza neanche la garanzia implicita\ndi NEGOZIABILITA' o APPLICABILITA' per un particolare scopo. Si veda la licenza generale pubblica GNU per maggiori dettagli.\r\n\r\nDovrebbe aver ricevuto una copia assieme a questo programma; se così non fosse, scriva alla Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA."
IDS_LICENSE "Este programa é software livre; você pode redistribuí-lo e/ou modificá-lo sob os termos da GNU General Public License conforme publicada pela Free Software Foundation; quer a versão 2 da licença, ou (conforme você escolha) qualquer versão posterior.\r\n\r\nEste programa é distribuído com a esperança de que seja útil, mas SEM QUALQUER GARANTIA; mesmo sem a garantia implícita de MERCANTIBILIDADE OU ADEQUAÇÃO A UM DETERMINADO PROPÓSITO. Para mais detalhes, veja a GNU General Public License.\r\n\r\nVocê deve ter recebido uma cópia da GNU General Public License juntamente com este programa; caso contrário, escreva para a Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA."
IDS_LICENSE "Această aplicație este publică; fiind permisă modificarea și/sau (re)distribuția sa în termenii Licenței Publice Generale GNU publicată de Free Software Foundation; sau versiunea 2 a Licenței, sau (la alegere) a oricărei versiuni ulterioare.\r\n\r\nAceastă aplicație este distribuită doar în speranța de a fi utilă, FĂRĂ însă NICI O GARANȚIE; nici măcar cu garanția implicită a VANDABILITĂȚII sau a UTILITĂȚII ÎNTR-UN SCOP ANUME. Pentru mai multe detalii consultați Licența Publică Generală GNU.\r\n\r\nPuteți vedea această licență aici:\r\nhttp://www.gnu.org/licenses/gpl.html\r\n\r\nDe asemenea puteți consulta traduceri neoficiale ale acestei licențe aici:\r\nhttp://www.gnu.org/licenses/translations.html"
IDS_LICENSE, "Tento program je voľný softvér; môžete ho šíriť a/alebo modifikovať podľa podmienok GNU Všeobecnej verejnej licencie (GNU General Public License) ako bola zverejnená nadáciou Free Software Foundation; buď verzie 2 tejto licencie, alebo (podľa Vášho uváženia) niektorej neskoršej verzie.\r\n\r\nTento program je distribuovaný v nádeji, že bude užitočný, avšak BEZ AKEJKOĽVEK ZÁRUKY; rovnako bez záruky PREDAJNOSTI alebo VHODNOSTI PRE URČITÝ ÚČEL. Pre viac detailov si pozrite GNU Všeobecnú verejnú licenciu (GNU General Public License).\r\n\r\nKópiu Všeobecnej verejnej licencie GNU ste mali dostať spolu s týmto programom; ak nie, napíšte si o ňu na Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA."
IDS_LICENSE "Tento program je voľný softvér; môžete ho šíriť a/alebo modifikovať podľa podmienok GNU Všeobecnej verejnej licencie (GNU General Public License) ako bola zverejnená nadáciou Free Software Foundation; buď verzie 2 tejto licencie, alebo (podľa Vášho uváženia) niektorej neskoršej verzie.\r\n\r\nTento program je distribuovaný v nádeji, že bude užitočný, avšak BEZ AKEJKOĽVEK ZÁRUKY; rovnako bez záruky PREDAJNOSTI alebo VHODNOSTI PRE URČITÝ ÚČEL. Pre viac detailov si pozrite GNU Všeobecnú verejnú licenciu (GNU General Public License).\r\n\r\nKópiu Všeobecnej verejnej licencie GNU ste mali dostať spolu s týmto programom; ak nie, napíšte si o ňu na Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA."
IDS_LICENSE "Detta programmet är fri programvara; du kan distribuera det och/eller ändra det under villkoren enligt GNU General Public License som är utgivet av Free Software Foundation; version 2, eller (om du så önskar) alla senare versioner.\r\n\r\nDette programmet er utgivet i hopp om att det skall kunne användas, men DET FINNS INGA GARANTIER; även utan underförstådd garanti om SÄLJBARHET eller LÄMPLIGHET FÖR ETT VISST ÄNDAMÅL. Se på GNU General Public Licensen för mer detaljer.\r\n\r\nDu skall ha motagit en kopia av GNU General Public Licensen tillsammans med detta programmet; om du inte har fått det, skriv til Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA."
text <EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><20><><20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><20><><20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>.\n"
/INSERT Включва режим на вмъкване.\n\
/OVERSTRIKE Изключва режима на вмъкване.\n\
/EXENAME=именаexe Указва името на приложението за преглед/промяна на дневника\n\
и макротата. По подразбиране е cmd.exe.\n\
/HISTORY Показва дневника на заповедите.\n\
/LISTSIZE=размер Указва размера на дневника за заповеди.\n\
/REINSTALL Изчиства дневника на заповедите.\n\
/MACROS Показва зададените макрота.\n\
/MACROS:именаexe Показва зададените макрота за определено приложение.\n\
/MACROS:ALL Показва зададените макрота за всички приложения.\n\
/MACROFILE=файловоиме Зарежда макрота от файл.\n\
macroname Указва името на макрото, което да бъде създадено.\n\
/C Skai<EFBFBD>iuoti eilutes, kuriose yra ie<EFBFBD>komas tekstas\n\
/I Ignoruoti raid<EFBFBD>i<EFBFBD> dyd<EFBFBD>\n\
/C Skaičiuoti eilutes, kuriose yra ieškomas tekstas\n\
/I Ignoruoti raidžių dydį\n\
/N Numeruoti vaizduojamas eilutes, pradedant nuo 1\n\
/V Spausdinti eilutes, kuriose n<EFBFBD>ra ie<EFBFBD>komo teksto"
/V Spausdinti eilutes, kuriose nėra ieškomo teksto"
IDS_NO_SUCH_FILE, "FIND: %s: Tokios bylos n<EFBFBD>ra\n"
IDS_NO_SUCH_FILE, "FIND: %s: Tokios bylos nėra\n"
IDS_CANNOT_OPEN, "FIND: %s: Nepavyko atverti bylos\n"
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.
