Ros-dev March 2010

ros-dev@reactos.org

22 participants
31 discussions

Re: [ros-dev] [ros-diffs] [sir_richard] 46103: [WIN32K]: Stop memory corruption when InstalledDisplayDrivers has more than one driver in the list. Note that driver loading is inherently broken right now, as the list of drivers is not parsed properly (this breaks eVb's VGA/VBE driver).
by Timo Kreuzer 11 Mar '10

11 Mar '10

Hi, I have a pending rewrite of the driver loading code in branches/reactos-yarotows. It should hopefully fix the issues and load eVb's driver. It's mostly done, I experienced some random lock ups, though and had to take a break from that code. Regards, Timo sir_richard(a)svn.reactos.org schrieb: > Author: sir_richard > Date: Thu Mar 11 18:46:15 2010 > New Revision: 46103 > > URL: http://svn.reactos.org/svn/reactos?rev=46103&view=rev > Log: > [WIN32K]: Stop memory corruption when InstalledDisplayDrivers has more than one driver in the list. Note that driver loading is inherently broken right now, as the list of drivers is not parsed properly (this breaks eVb's VGA/VBE driver). > > Modified: > trunk/reactos/subsystems/win32/win32k/objects/device.c > > Modified: trunk/reactos/subsystems/win32/win32k/objects/device.c > URL: http://svn.reactos.org/svn/reactos/trunk/reactos/subsystems/win32/win32k/ob… > ============================================================================== > --- trunk/reactos/subsystems/win32/win32k/objects/device.c [iso-8859-1] (original) > +++ trunk/reactos/subsystems/win32/win32k/objects/device.c [iso-8859-1] Thu Mar 11 18:46:15 2010 > @@ -59,12 +59,37 @@ > return TRUE; > } > > + > +NTSTATUS > +NTAPI > +EnumDisplayQueryRoutine(IN PWSTR ValueName, > + IN ULONG ValueType, > + IN PVOID ValueData, > + IN ULONG ValueLength, > + IN PVOID Context, > + IN PVOID EntryContext) > +{ > + if ((Context == NULL) && ((ValueType == REG_SZ) || (ValueType == REG_MULTI_SZ))) > + { > + *(PULONG)EntryContext = ValueLength; > + } > + else > + { > + DPRINT1("Value data: %S %d\n", ValueData, ValueLength); > + RtlCopyMemory(Context, ValueData, ValueLength); > + } > + > + return STATUS_SUCCESS; > +} > + > static BOOL FASTCALL > FindDriverFileNames(PUNICODE_STRING DriverFileNames, ULONG DisplayNumber) > { > RTL_QUERY_REGISTRY_TABLE QueryTable[2]; > UNICODE_STRING RegistryPath; > NTSTATUS Status; > + PWCHAR DriverNames = NULL; > + ULONG Length = 0; > > if (! GetRegistryPath(&RegistryPath, DisplayNumber)) > { > @@ -73,23 +98,40 @@ > } > > RtlZeroMemory(QueryTable, sizeof(QueryTable)); > - QueryTable[0].Flags = RTL_QUERY_REGISTRY_REQUIRED | RTL_QUERY_REGISTRY_DIRECT; > + QueryTable[0].Flags = RTL_QUERY_REGISTRY_REQUIRED | RTL_QUERY_REGISTRY_NOEXPAND; > QueryTable[0].Name = L"InstalledDisplayDrivers"; > - QueryTable[0].EntryContext = DriverFileNames; > + QueryTable[0].EntryContext = &Length; > + QueryTable[0].QueryRoutine = EnumDisplayQueryRoutine; > > Status = RtlQueryRegistryValues(RTL_REGISTRY_ABSOLUTE, > RegistryPath.Buffer, > QueryTable, > NULL, > NULL); > + // DPRINT1("Status: %lx\n", Status); > + if (Length) > + { > + DriverNames = ExAllocatePool(PagedPool, Length); > + // DPRINT1("Length allocated: %d\n", Length); > + Status = RtlQueryRegistryValues(RTL_REGISTRY_ABSOLUTE, > + RegistryPath.Buffer, > + QueryTable, > + DriverNames, > + NULL); > + if (!NT_SUCCESS(Status)) DriverNames = NULL; > + } > + > ExFreePoolWithTag(RegistryPath.Buffer, TAG_RTLREGISTRY); > if (! NT_SUCCESS(Status)) > { > DPRINT1("No InstalledDisplayDrivers value in service entry found\n"); > return FALSE; > } > - > - DPRINT("DriverFileNames %S\n", DriverFileNames->Buffer); > + > + RtlInitUnicodeString(DriverFileNames, DriverNames); > + DriverFileNames->Length = Length; > + DriverFileNames->MaximumLength = Length; > + //DPRINT1("DriverFileNames %wZ\n", DriverFileNames); > > return TRUE; > } > @@ -301,7 +343,7 @@ > continue; > } > > - DPRINT("Display driver %S loaded\n", CurrentName); > + DPRINT1("Display driver %S loaded\n", CurrentName); > > ExFreePoolWithTag(DriverFileNames.Buffer, TAG_RTLREGISTRY); > > > > >

1 0

[ros-diffs] [tkreuzer] 46089: [KMTYPES] Add a compilation hack for gcc
by Dmitry Gorbachev 11 Mar '10

11 Mar '10

Hi! typedef struct _TRANSACTION_NOTIFICATION_RECOVERY_ARGUMENT { GUID EnlistmentId; #if !defined(__cplusplus) UOW UOW; #else ::UOW UOW; #endif } TRANSACTION_NOTIFICATION_RECOVERY_ARGUMENT, *PTRANSACTION_NOTIFICATION_RECOVERY_ARGUMENT; can be used instead of a hack.

1 0

[ros-diffs] [cwittich] 46076: [PSDK] add extern GUID_NULL
by James Tabor 11 Mar '10

11 Mar '10

Hi! The ./psdk/cguid.h:26:extern const IID GUID_NULL; has it already! Maybe include it in the test rbuild file? James

1 0

Google summer code
by Mgr. Janusz Chmiel 10 Mar '10

10 Mar '10

Dear core developers of Reactos, I would like to inform YOu about my findings related to screen readers support while using Wine emulator, because as i understood, Wine and Reactos are sharing crucial information to make those tools as much Windows compatible as it is possible. 1. NVDA is not able to run in The Wine, which can be downloaded from PCLINUXOSS repositories, NVDA is only plaing a welcome gong and then i Am receiving series of error messages, one related to winmm.dll and also other related to MSAA support. oleacc.dll very probably. When i tried one Czech very little screen reader based on using API hooks and API calls, program is written by using Microsoft Visual Basic 6.0 professional edition and crucial helper dlls are written in Microsoft visual C++ but not Express 2005 edition, older release of this programming language has been used. What are mi findings? Algorithms used for generating synthesizet speech are working ammazingly in Wine. but other algorithms for fetching names of buttons, objects and for getting textual information are not working. But i Am afraid, that implementing reliable MSAA support is so much complex for core developers of Wine or Reactos, that it will never been possible to reach this goal. Many of us are very intelligent and professional developers, who are even able to use Assembler while compiling and developing Reactos core components, but i think, that without access to Windows MSAA source code developing reliable MSAA support will be only big dream, which would never been come true. Reactos is in general very stable and reliable operating system and i believe, that in future, also many device drivers will work with Reactos. If also MSAA support would be added in future, this would be very positive for many handicapped users, because MSAA is being used not only by screen readers, but also while using other assistive technology, which are based on MSAA calls. BUt i think, that Microsoft will not want to reveal The MSAA source code, the modified source code of oleacc.dll, user32.dll and others. Accessing source code is fortunately possible, but i read about this, that some fee must be paied to Microsoft to get access to their source code repository, and i think, that there will be very restricted conditions.

2 1

Re: [ros-dev] [ros-diffs] [evb] 45874: - New Framebuffer (Linear) Display Driver to support new unified VGA/VBE miniport. Based on NT4 DDK Sample, with modifications by me (marked with // eVb) to support new functionality needed for 2003-era driver. - Also use
by Steven Edwards 10 Mar '10

10 Mar '10

On Fri, Mar 5, 2010 at 12:29 PM, <evb(a)svn.reactos.org> wrote: > - New Framebuffer (Linear) Display Driver to support new unified VGA/VBE miniport. Based on NT4 DDK Sample, with modifications by me (marked with // eVb) to support new functionality needed for 2003-era driver. > - Also used Virtual Box Display Driver as sample, which is based on "GPL" Windows 2003 DDK sample driver. Could not use 2003 DDK sample directly because of licensing issues, and feel unsafe about VirtualBox "GPL" driver that says "PATENTED AND ONLY FOR USE IN MICROSOFT PRODUCTS". > - Note that old driver was based off DDK sample too, but with variables renamed (some comments identical!) and code reformatted, then marked as "GPL". This is not very good way to share/use code... one day someone can teach you lesson. I am not sure what this comment is supposed to mean. Are you saying that the wishes of the Author should not be respected? The PATENTED statement is an extra qualifier. We are talking about two issues, patents (which are vague and Copyright which is clear). Clearly the author being Microsoft does not want it's code used as a derived work for a non-windows OS. As the copyright owner and licensor that is their right. Or do you propose that developers ignore the copyright law and do it anyway? If you want to import the Windows 2003 DDK driver and create a derived work in violation of the terms, do the rest of us a favor and upload a scanned copy of your drivers license, passport or whatever as a resource for this driver so Microsoft will know exactly who to sue. -- Steven Edwards "There is one thing stronger than all the armies in the world, and that is an idea whose time has come." - Victor Hugo

2 2

Re: [ros-dev] [ros-diffs] [sir_richard] 46042: [KERNEL32/CSRSS]: Register new Win32 threads with CSRSS. Add a bunch of lookup, creation, allocation, hashing and CSR thread management routines from the unused CSRSRV in trunk. [CSRSS]: Bang in the new thread support
by Gabriel ilardi 09 Mar '10

09 Mar '10

Hi, There's a regression that I narrowed to this range: 46028 (no regression) - 46043 (regression). Tested in VirtualBox, in 3rd stage you get: (base/services/umpnpmgr/umpnpmgr.c:2132) CreateProcessAsUserW failed with error 1375 (ERROR_TOKEN_ALREADY_IN_USE), and no new hardware wizard prompt. Regards, Gabriel. Here's the full debug log for 3rd stage: (ntoskrnl/kd/kdio.c:297) ----------------------------------------------------- (ntoskrnl/kd/kdio.c:298) ReactOS 0.4-SVN (Build 20100309-r46043) (ntoskrnl/kd/kdio.c:299) Command Line: /DEBUG /DEBUGPORT=COM1 /BAUDRATE=115200 /SOS (ntoskrnl/kd/kdio.c:303) ARC Paths: multi(0)disk(0)rdisk(0)partition(1) \ multi(0)disk(0)rdisk(0)partition(1) \ReactOS\ (ntoskrnl/ke/i386/cpu.c:746) Prefetch Cache: 64 bytes L2 Cache: 524288 bytes L2 Cache Line: 64 bytes L2 Cache Associativity: 16 WARNING: HalStopProfileInterrupt at hal/halx86/generic/profil.c:24 is UNIMPLEMENTED! (ntoskrnl/mm/mminit.c:289) 0x80000000 - 0x80C00000 Boot Loaded Image (ntoskrnl/mm/mminit.c:293) 0x80C00000 - 0x87000000 Paged Pool (ntoskrnl/mm/mminit.c:297) 0xB0000000 - 0xB0300000 PFN Database (ntoskrnl/mm/mminit.c:301) 0xB0300000 - 0xB0AC8000 ARM³ Non Paged Pool (ntoskrnl/mm/mminit.c:305) 0xBC000000 - 0xBD000000 System View Space (ntoskrnl/mm/mminit.c:309) 0xBD000000 - 0xC0000000 Session Space (ntoskrnl/mm/mminit.c:312) 0xC0000000 - 0xC0300000 Page Tables (ntoskrnl/mm/mminit.c:315) 0xC0300000 - 0xC0400000 Page Directories (ntoskrnl/mm/mminit.c:318) 0xC0400000 - 0xC0800000 Hyperspace (ntoskrnl/mm/mminit.c:322) 0xE1000000 - 0xED400000 ARM³ Paged Pool (ntoskrnl/mm/mminit.c:325) 0xF4C00000 - 0xFA305000 System PTE Space (ntoskrnl/mm/mminit.c:328) 0xFA305000 - 0xFFBE0000 Non Paged Pool Expansion PTE Space (ntoskrnl/ke/i386/kiinit.c:53) Large Page support detected but not yet taken advantage of (ntoskrnl/ke/i386/patpge.c:45) Global page support detected but not yet taken advantage of (ntoskrnl/ke/i386/patpge.c:58) PAT support detected but not yet taken advantage of (ntoskrnl/ke/i386/mtrr.c:24) MTRR support detected but not yet taken advantage of PC Compatible Eisa/Isa HAL has been initialized WARNING: IoReportResourceUsage at ntoskrnl/io/iomgr/iorsrce.c:700 is UNIMPLEMENTED! WARNING: IoReportResourceUsage at ntoskrnl/io/iomgr/iorsrce.c:700 is UNIMPLEMENTED! (ntoskrnl/io/iomgr/driver.c:1392) '\Driver\BUSLOGIC' initialization failed, status (0xc00000c0) (drivers/network/ndis/ndis/miniport.c:2413)(NdisMRegisterMiniport) Initializing an NDIS 5.0 miniport (drivers/network/ndis/ndis/config.c:485)(NdisReadConfiguration) ZwQueryValueKey #1 failed for NetworkAddress, status 0xc0000034 (drivers/network/ndis/ndis/config.c:700)(NdisReadNetworkAddress) NdisReadConfiguration failed (c0000001) (drivers/network/ndis/ndis/efilter.c:96)(EthFilterDprIndicateReceive) Filter is NULL (drivers/network/ndis/ndis/efilter.c:96)(EthFilterDprIndicateReceive) Filter is NULL (drivers/network/ndis/ndis/efilter.c:96)(EthFilterDprIndicateReceive) Filter is NULL (drivers/network/ndis/ndis/efilter.c:129)(EthFilterDprIndicateReceiveComplete) Filter is NULL (ntoskrnl/io/iomgr/deviface.c:1005) IoGetDeviceObjectPointer() failed with status 0xc0000010 wood_uhci: Entering DriverEntry(), RegistryPath= \Registry\Machine\System\CurrentControlSet\Services\usbdriver (ntoskrnl/io/iomgr/driver.c:1392) '\Driver\usbdriver' initialization failed, status (0xc0000001) Unhandled event type: 6 (ntoskrnl/io/iomgr/file.c:420) Using IopParseDevice() hack. Requested invalid attributes: 1 (ntoskrnl/io/iomgr/file.c:420) Using IopParseDevice() hack. Requested invalid attributes: 1 Boot took 14375646481 cycles! Interrupts: 1079 System Calls: 4850 Context Switches: 291 (subsystems/win32/win32k/ntuser/sysparams.c:1499) UserSystemParametersInfo called without active windowstation. err:(dll/win32/user32/misc/dllmain.c:367) GetConnected (subsystems/win32/win32k/ntuser/desktop.c:604) RtlQueryRegistryValues failed for PaintDesktopVersion (c0000034) (subsystems/win32/win32k/ntuser/sysparams.c:1499) UserSystemParametersInfo called without active windowstation. (subsystems/win32/win32k/objects/gdiobj.c:1108) Attempted to lock object 0x0 of wrong type (Handle: 0x0, requested: 0x50000) (subsystems/win32/win32k/objects/gdiobj.c:1108) Attempted to lock object 0x0 of wrong type (Handle: 0x0, requested: 0x50000) (subsystems/win32/win32k/objects/gdiobj.c:1108) Attempted to lock object 0x0 of wrong type (Handle: 0x0, requested: 0x50000) (subsystems/win32/win32k/objects/gdiobj.c:1108) Attempted to lock object 0x0 of wrong type (Handle: 0x0, requested: 0x50000) err:(dll/win32/user32/windows/window.c:281) CreateWindowExA RegisterSystemControls Boot took 21134116768 cycles! Interrupts: 1632 System Calls: 8466 Context Switches: 1260 (subsystems/win32/win32k/ntuser/desktop.c:604) RtlQueryRegistryValues failed for PaintDesktopVersion (c0000034) err:(dll/win32/advapi32/service/sctrl.c:668) ScmrSetServiceStatus() failed (Error 1722) err:(dll/win32/advapi32/service/sctrl.c:668) ScmrSetServiceStatus() failed (Error 1722) Boot took 24388307864 cycles! Interrupts: 1753 System Calls: 11824 Context Switches: 1859 err:(dll/win32/advapi32/service/sctrl.c:668) ScmrSetServiceStatus() failed (Error 1722) err:(dll/win32/advapi32/service/sctrl.c:668) ScmrSetServiceStatus() failed (Error 1722) (lib/rtl/actctx.c:874) Unsupported yet language attribute (*) (lib/rtl/actctx.c:2007) looking for name=Microsoft.Windows.Common-Controls version=6.0.0.0 arch=X86 Using shell hooks for notification of shell events. (subsystems/win32/win32k/ntuser/hook.c:1202) Not implemented: HookId 3 Global TRUE fixme:(dll/win32/kernel32/file/backup.c:35) BackupRead is UNIMPLEMENTED! fixme:(dll/win32/kernel32/file/backup.c:35) BackupRead is UNIMPLEMENTED! (lib/rtl/actctx.c:874) Unsupported yet language attribute (*) (lib/rtl/actctx.c:2007) looking for name=Microsoft.Windows.Common-Controls version=6.0.0.0 arch=X86 CreateDirectory("C:\Documents and Settings\Administrator.REACTOS\Application Data\Microsoft") CreateDirectory("C:\Documents and Settings\Administrator.REACTOS\Application Data\Microsoft\Internet Explorer") (ntoskrnl/se/semgr.c:294) SidInToken Calls: 10000 CreateDirectory("C:\Documents and Settings\Administrator.REACTOS\Application Data\Microsoft\Internet Explorer\Quick Launch") (base/services/umpnpmgr/umpnpmgr.c:2132) CreateProcessAsUserW failed with error 1375 (base/services/umpnpmgr/umpnpmgr.c:2132) CreateProcessAsUserW failed with error 1375 (base/services/umpnpmgr/umpnpmgr.c:2132) CreateProcessAsUserW failed with error 1375 (base/services/umpnpmgr/umpnpmgr.c:2132) CreateProcessAsUserW failed with error 1375 _________________________________________________________________ Più spazio per le tue esigenze. Hotmail va oltre i 5GB http://www.windowslive.it/hotmail/SpazioDisponibile.aspx

1 0

Re: [ros-dev] [ros-diffs] [jimtabor] 46006: - [User32_winetest] - Win : Remove test_capture from service. This is related to TrackMouseEvent issues which use SetCapture.
by Aleksey Bragin 09 Mar '10

09 Mar '10

I was blamed for doing this the other day. So do we skip tests or don't? If yes, they must be marked as skipped, not silently ignored. WBR, Aleksey Bragin. On Mar 8, 2010, at 11:52 PM, jimtabor(a)svn.reactos.org wrote: > Author: jimtabor > Date: Mon Mar 8 21:52:04 2010 > New Revision: 46006 > > URL: http://svn.reactos.org/svn/reactos?rev=46006&view=rev > Log: > - [User32_winetest] > - Win : Remove test_capture from service. This is related to > TrackMouseEvent issues which use SetCapture. > > Modified: > trunk/rostests/winetests/user32/win.c > > Modified: trunk/rostests/winetests/user32/win.c > URL: http://svn.reactos.org/svn/reactos/trunk/rostests/winetests/ > user32/win.c?rev=46006&r1=46005&r2=46006&view=diff > ====================================================================== > ======== > --- trunk/rostests/winetests/user32/win.c [iso-8859-1] (original) > +++ trunk/rostests/winetests/user32/win.c [iso-8859-1] Mon Mar 8 > 21:52:04 2010 > @@ -5979,7 +5979,7 @@ > test_capture_1(); > test_capture_2(); > test_capture_3(hwndMain, hwndMain2); > - test_capture_4(); > +// test_capture_4(); > > test_CreateWindow(); > test_parent_owner(); > >

1 0

ros-diffs] [sir_richard] 46027: [WIN32K]: Stub-plement NtUserSetInformationThread for UserInitiateShutdown and UserEndShutdown. I think we're supposed to notify Winlogon at some point, and check for races. Part of new shutdown code.
by James Tabor 09 Mar '10

09 Mar '10

Hi, Look for ntuser/defwnd.c and read the beginning of the file that deals with shutdown messages, it passes all the wine tests. If you need something, just ask, I will implement it. Thanks, James

1 0

Re: [ros-dev] [ros-diffs] [cgutman] 46011: [CMBATT] - Add initial cmbatt driver which mostly stubs - All it does right now is register a battery device with the battc driver [BATTC] - Add a nearly complete battc driver (the only stuff missing is WMI) - Tested with
by Ged Murphy 09 Mar '10

09 Mar '10

cgutman(a)svn.reactos.org wrote: > +BCLASSAPI > +NTSTATUS > +DDKAPI > +BatteryClassSystemControl(PVOID ClassData, > + PWMILIB_CONTEXT WmiLibContext, > + PDEVICE_OBJECT DeviceObject, > + PIRP Irp, > + PSYSCTL_IRP_DISPOSITION Disposition) > +{ > + NTSTATUS Status; > + > + UNIMPLEMENTED > + > + /* FIXME: Uncomment when WmiCompleteRequest is implemented */ > +#if 0 > + Status = STATUS_WMI_GUID_NOT_FOUND; > + WmiCompleteRequest(DeviceObject, > + Irp, > + Status, > + 0, > + IO_NO_INCREMENT); > +#else > + Irp->IoStatus.Status = Status = STATUS_WMI_GUID_NOT_FOUND; > + Irp->IoStatus.Information = 0; > + > + IoCompleteRequest(Irp, IO_NO_INCREMENT); > +#endif > + > + return Status; > +} > + This isn't how WMI works at all

1 0

Trunk lockage
by Aleksey Bragin 09 Mar '10

09 Mar '10

Hi, trunk is locked until (at least) debug buildslave is back online. Also, it's a good occasion to actually FIX all those hanging tests (fix, hackfix, anyotherwaytofix) which lead to buildslaves timewaste (compare: 4 minutes to build and 1 hour to test). This is unbearable. Test results should be available within minutes, and allow easy tracking of failures. WBR, Aleksey Bragin.

9 9

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Ros-dev March 2010