Why doesn't mirror driver catch BitBlt sometimes?

TheMSsForum.com: The Microsoft Software Forum

The MSS Forum ‹ Drivers
- Archive
  - Biz
  - MCSE
  - CRM
  - Drivers
  - Framework
  - ADO
  - ASP
  - Compact
  - Forms
  - Dotnet
  - C#
  - VB
  - FontpageGen
  - Excel
  - WorkSheet
  - Exchange
  - Setup
  - Fox
  - Fontpage
  - ASP
  - IIS
  - Entourage
  - Money
  - Messanger
  - PocketPC
  - Powerpoint
  - Project
  - Publisher
  - Excel
  - VB
  - Security
  - Portal
  - Services
  - SQLServerDev
  - SVCS
  - SQLServer
  - VB
  - VC
  - MFC
  - ExcelGen

- Previous
  - 1
    - WMI Property Page for Driver I have added WMI support for the driver and I can access the class and the properties using wbemtest and other utlilities. However, there is something that the DDK mentions about having a Property Page and I am not able to see where the propety page comes or how to enable the property page ? I have browed theough the device manager, services menu et al but I cant find it. this is what the DDK says... "Drivers can automatically expose any WMI classes they implement on their property sheet by using the WMI generic property page provider." I have enabled everything through the INF that the DDK says about the property page provider and the co-installer as below, but I still can't find the damn property page..neither in the device manger( by right clicking the device or elsewhere). Any help is appreciated. Thanks. This is from the DDK and I am doing everything as below. Enabling the Generic Property Page Provider To use the WMI generic property page provider with your driver, you must enable wmiprop.dll as a device co-installer. In the CoInstaller add-registry-section, register wmiprop.dll as a co-installer for the WMI class by adding a value entry for the class GUID under the HKLM\System\CurrentControlSet\Control\CoDeviceInstallers key. The value for the value entry is "WmiProp.dll, WmiPropCoInstaller". For example: ; This section is defined in the CoInstaller section, as follows. ; [CoInstaller] ; AddReg = CoInstaller_AddReg [CoInstaller_AddReg] HKLM, System\CurrentControlSet\Control\CoDeviceInstallers, ClassGUID, 0x00010000, "WmiProp.dll, WmiPropCoInstaller" ClassGUID is the GUID for the WMI class. See Registering a Class Co-installer for details. You must also specify the particular WMI classes to be exposed through the generic property provider. To do this, set the WmiConfigClasses value-entry to be a comma-separated list of the WMI classes in the add-registry-section of the device class or device hardware instance. ; the device class AddReg section. [device_class_AddReg] HKR,,"WmiConfigClasses",0x00000000,"class1,class2" ; the device hardware instance AddReg section. [device_hw_inst_AddReg] HKR,,"WmiConfigClasses",0x00000000,"class3" Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79511
  - 2
    - Pen Drive Ids As I obtain this information of pen drive? (vendorid, productid...) Use XP SP2 and VB.net 2005 Some idea? Thanks! Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79510
  - 3
    - Help with Hidd_GetAttributes The Hidd_GetAttributes function does not return the information to me of my device USB (they pen drive). It returns all 0. GetLastError returns code 50. Use XP SP2 and VB.net 2005 Some idea? Thanks! PS: CreateFile works OK Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79502
  - 4
    - Driver project configuration. Hi all, I'm interested in developing Win32 device drivers using the Visual Studio .NET 2003 IDE. (I'm purposely trying to avoid using DDK make files for internal maintenance reasons - trust me, they warrant this action.) Since I'm developing the drivers from scratch I thought I should be able to do this without problems. I've read a few articles about device driver development and have successfully implemented an example driver using the IDE by manually specifying the necessary compiler switches, but I'm not sure how to make certain features work. For example, if I enable C++ exception handling, I receive the following link error: Defs.obj : error LNK2019: unresolved external symbol __CxxThrowException@8 referenced in function "void * __cdecl operator new(unsigned int,enum EMemPool)" (??2@YAPAXIW4EMemPool@@@Z) I discovered that if I manually link in the "libcmt.lib" library then the symbol is resolved, but this is no good because it results in the following link errors (as I expected): kernel32.lib(KERNEL32.dll) : warning LNK4237: /SUBSYSTEM:NATIVE specified when importing from 'KERNEL32.DLL'; Use /SUBSYSTEM:CONSOLE or /SUBSYSTEM:WINDOWS. LINK : warning LNK4078: multiple '.rdata' sections found with different attributes (C0000040) My project is obviously implemented in C++. Now, I know just about every driver developer out there is thinking that one should never use C++ in drivers, but I have yet to discover any compelling reason why, so unless you've got one, please don't respond to this post only to bark at me about how C++ is evil (or whatever.) Please understand that when you respond to the post others that have the answers I need won't because they will assume I have received the information I need. I've overridden the global new and delete operators so I can control whether memory is allocated from the paged or non-paged pool and I've carefully selected my compiler switches to insure that all necessary sections exist and have the proper attributes. At this time it seems that the only problem I have is that I'm missing the VC++ exception handling routines that live the CRT libraries. So I either have to not use exceptions, which is a major loss to me, or find out how to resolve these symbols. Can anyone point me in the right direction? I'm also not sure what I'll be loosing by opting not to use the DDK build system (make files) and using the VS.NET 2003 IDE instead. Is there anything or do the make files primarily just configure build switches, etc? Thanks in advance, Matt Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79501
  - 5
    - Mount manager and virtual disk drive Hello, I devloped a virtual disk driver based on the ramdisk sample several years ago. Now we found out, that I can't share the drives on the network or defrag it. Every time I try create a share name I get an error like "device or directory dosen't exist". I guess that IoRegisterDeviceInterface () may solve my problem, but I don't have a PDO... Can swenum supply me a "virtual PDO" ? Is there anywhere a sample for swenum? Regards Kalle Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79495
  - 6
    - KSDATARANGE problem (AVSTREAM) Hello. i defined a several struct describing output pin resolution of AVSTREM minidriver like this #define D_X 160 #define D_Y 120 const KS_DATARANGE_VIDEO FormatRGB24Bpp_Capture0 = { // KSDATARANGE { sizeof (KS_DATARANGE_VIDEO), // FormatSize 0, // Flags (Ignored) D_X * D_Y * 3, // SampleSize (Ignored) 0, // Reserved (Reserved) STATICGUIDOF( KSDATAFORMAT_TYPE_VIDEO ), // aka. MEDIATYPE_Video 0xe436eb7d, 0x524f, 0x11ce, 0x9f, 0x53, 0x00, 0x20, 0xaf, 0x0b, 0xa7, 0x70, //MEDIASUBTYPE_RGB24, STATICGUIDOF( KSDATAFORMAT_SPECIFIER_VIDEOINFO ) // aka. FORMAT_VideoInfo }, TRUE, // BOOL, bFixedSizeSamples (all samples same size?) TRUE, // BOOL, bTemporalCompression (all I frames?) KS_VIDEOSTREAM_CAPTURE,// Reserved (was StreamDescriptionFlags) 0, // Reserved (was MemoryAllocationFlags (KS_VIDEO_ALLOC_*)) // _KS_VIDEO_STREAM_CONFIG_CAPS { STATICGUIDOF( KSDATAFORMAT_SPECIFIER_VIDEOINFO ), // GUID KS_AnalogVideo_None, D_X, D_Y, // InputSize, (the inherent size of the incoming signal // with every digitized pixel unique) D_X, D_Y, // MinCroppingSize, smallest rcSrc cropping rect allowed D_X, D_Y, // MaxCroppingSize, largest rcSrc cropping rect allowed 1, // CropGranularityX, granularity of cropping size 1, // CropGranularityY 1, // CropAlignX, alignment of cropping rect 1, // CropAlignY; D_X, D_Y, // MinOutputSize, smallest bitmap stream can produce D_X, D_Y, // MaxOutputSize, largest bitmap stream can produce 1, // OutputGranularityX, granularity of output bitmap size 1, // OutputGranularityY; 0, // StretchTapsX 0, // StretchTapsY 0, // ShrinkTapsX 0, // ShrinkTapsY 333333, // MinFrameInterval, 100 nS units 10000000, // MaxFrameInterval, 100 nS units 8 * 3 * 1 * D_X * D_Y, // MinBitsPerSecond; 8 * 3 * 30 * D_X * D_Y // MaxBitsPerSecond; }, // KS_VIDEOINFOHEADER { 0,0,0,0, // RECT rcSource; 0,0,0,0, // RECT rcTarget; D_X * D_Y * 8 * 3 * 30, // DWORD dwBitRate; 0L, // DWORD dwBitErrorRate; 333333, // REFERENCE_TIME AvgTimePerFrame; // KS_BITMAPINFOHEADER sizeof (KS_BITMAPINFOHEADER), // DWORD biSize; D_X, // LONG biWidth; D_Y, // LONG biHeight; 1, // WORD biPlanes; 24, // WORD biBitCount; KS_BI_RGB, // DWORD biCompression; D_X * D_Y * 3, // DWORD biSizeImage; 0, // LONG biXPelsPerMeter; 0, // LONG biYPelsPerMeter; 0, // DWORD biClrUsed; 0 // DWORD biClrImportant; } }; #undef D_X #undef D_Y And found one problem, when i defined a several resolutions with 160 width (for i.e. 160x120, 160x128, 160x140) AMCAP incorrectly draws resolution list in stream format dialog resolution list - some items is on the incorerect position (i suppose that they should appear in the order as i define it in KSPIN_DESCRIPTOR_EX) and another bug is when i select 160x120 and then opens this dialog again the current selection is 160x140 :( Can anybody advise what the problem is? Thanks. -Tulup Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79488
  - 7
    - Which class to use? Hi to all, i am planning to start developing a usb device which has to send continious numbers to the application. I don't want to write a driver. Can i use the drivers already existing. If so which device class can i use for that. Pls shre ur knowledge with so that i can finish my mini project by capturing continious data from usb. Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79484
  - 8
    - Error in calling I2COpen function Hi, I've modified WDM sample code ATIShare source code to access the I2C bus. With the code, given below, I am able to get I2COpen & I2CAccess function pointers from Video Adapter. But I2COpen function always returns 0xC00000C2 error code. It is defined in ntstatus.h as #define STATUS_ADAPTER_HARDWARE_ERROR ((NTSTATUS)0xC00000C2L) Has anyone experienced this problem before? There is no documentation available for this. Any help is appreciated. I'm calling LockI2Cprovider function after LocateAttachI2CProvider function call. The program fails in LockI2Cprovider () function. BOOLEAN LocateAttachI2CProvider( PDEVICE_OBJECT pVideoDeviceObject, PI2C_CONTEXT pstI2CContext ) { PIRP pIrp; BOOLEAN bResult = FALSE; PIO_STACK_LOCATION pNextStack; NTSTATUS ntStatus; KEVENT Event; I2CINTERFACE* pI2CInterface; if (pVideoDeviceObject == NULL || pstI2CContext == NULL) { return FALSE; } pI2CInterface = pstI2CContext->m_i2cProviderInterface; if (pI2CInterface == NULL) { return FALSE; } pIrp = IoAllocateIrp (pVideoDeviceObject->StackSize, FALSE); if( pIrp == NULL) { return FALSE; } pNextStack = IoGetNextIrpStackLocation (pIrp); // original if (pNextStack == NULL) { return FALSE; } pIrp->IoStatus.Status = STATUS_NOT_SUPPORTED; pNextStack->MajorFunction = (UCHAR)IRP_MJ_PNP; pNextStack->MinorFunction = IRP_MN_QUERY_INTERFACE; KeInitializeEvent( &Event, NotificationEvent, FALSE); IoSetCompletionRoutine ( pIrp, I2CScriptIoSynchCompletionRoutine, &Event, TRUE, TRUE, TRUE); pNextStack->Parameters.QueryInterface.InterfaceType = (struct _GUID *) &GUID_I2C_INTERFACE; pNextStack->Parameters.QueryInterface.Size = sizeof( I2CINTERFACE); pNextStack->Parameters.QueryInterface.Version = 1; pNextStack->Parameters.QueryInterface.Interface = (PINTERFACE) pI2CInterface; pNextStack->Parameters.QueryInterface.InterfaceSpecificData = NULL; ntStatus = IoCallDriver (pVideoDeviceObject, pIrp); if( ntStatus == STATUS_PENDING) { KeWaitForSingleObject( &Event, Suspended, KernelMode, FALSE, NULL); } if ((pI2CInterface->i2cOpen == NULL) || (pI2CInterface->i2cAccess == NULL)) { return FALSE; } else { DebugPrint ("deviceInfo->m_i2cProviderInterface->i2cOpen 0x%x", pI2CInterface->i2cOpen); DebugPrint ("deviceInfo->m_i2cProviderInterface->i2cAccess 0x%x", pI2CInterface->i2cAccess); } pstI2CContext->m_pdoDriver= pVideoDeviceObject; bResult = TRUE; if( pIrp != NULL) IoFreeIrp( pIrp); return bResult; } BOOLEAN LockI2CProvider (PI2C_CONTEXT pstContext) { NTSTATUS status; if ((pstContext->m_i2cProviderInterface->i2cOpen == NULL) || (pstContext->m_i2cProviderInterface->i2cAccess == NULL) || (pstContext->m_pdoDriver == NULL)) { return FALSE; } pstContext->i2cAccessBlock->Status = I2C_STATUS_NOERROR; status = pstContext->m_i2cProviderInterface->i2cOpen ( pstContext->m_pdoDriver, TRUE, pstContext->i2cAccessBlock); if (status != STATUS_SUCCESS) { DebugPrint ( "CI2CScript: LockI2CProvider() status = 0x%lx\n", status); return FALSE; } pstContext->m_dwI2CAccessKey = pstContext->i2cAccessBlock->dwCookie; DebugPrint ("LockI2CProvider m_dwI2CAccessKey %ld", pstContext->i2cAccessBlock->dwCookie); return TRUE; } Thanks VK Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79483
  - 9
    - how to get volume instance for a USB device ? Anyone has any suggestion how the interface ID in the volume class can be acquired based on a known USB drive ? A parameter in USB device registry, "ParantIdIndex", contains the right information, but there is no corresponding SPDRP_xxx for that parameter. Thanks -Gary Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79472
  - 10
    - Printing drivers architecture Hi I've read the information in the DDK about the printing architecture and I have several questions: 1. Is there a better documentation about the printing architecture then the MSDN? 2. Why some printers print in raw and some in EMF? How this decision is being taken after an application is issuing a "Print" command? Is there a way to always produce an EMF file instead of RAW, regardless the printer being used? 3. My goal is eventually to get an EMF file from each print request, analyze it and block / deny the printing. So, Where in the "chain" is the best place to enter to do such a thing? Thanks Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79463
  - 11
    - Printing drivers architecture Hi I've read the information in the DDK about the printing architecture and I have several questions: 1. Is there a better documentation about the printing architecture then the MSDN? 2. Why some printers print in raw and some in EMF? How this decision is being taken after an application is issuing a "Print" command? Is there a way to always produce an EMF file regardless the printer I'm using? 3. My goal is eventually to get an EMF file from each print request, analyze it and block / deny the printing. So, Where in the "chain" is the best place to enter to do such a thing? Thanks Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79462
  - 12
    - USB Selective suspend question again Is there a possibility of race condition when canceling USB idle request? From the WDK docum ( build 5356): "The following list indicates the 3 most important sets of circumstances in which a driver might cancel an idle request and specifies what the driver must do in each case: (1)The bus driver has not called the USB Idle Notification Callback Routine. The driver must use IoCancelIrp to cancel the idle request IRP. (2)The bus driver has called the USB idle notification callback routine. Drivers must not cancel an idle request IRP within the idle notification callback routine. (3) ... If the device is already in a low power state, it is too late to cancel the idle request IRP ... " Since the bus driver calls the callback at passive IRQL, another thread that tries to cancel the idle IRP needs some way to sync with it ? OTOH if the idle IRP can be canceled in parallel with the callback, but the driver needs to serialize something other (like sending D2 power irp), please say exactly so. Regards, Pavel Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79459
  - 13
    - Virtual Webcam followup, win98 install issues Hi I've successfully implemented a virtual webcam device driver which plays video files , based on the TESTCAP DDK sample. Its working perfectly on both Win2k/xp and Win98. However, while on 2K/XP , i can install the driver automatically using the sample utility provided in the DDK ( %BASEDIR%\SRC\SETUP\INSTALL ), Im having trouble with doing that on Win98. On Win98 im trying to use the CINSTALL utility. It recognizes the INF file and installs the sys files and device manager shows the entry but when i run AMCAP, I get an error "Could not create video capture filter : 0x80070002" aka File not found. Running a registry monitor shows that its failing to find some registry keys after which it throws the error. Using the Add hardware wizard works seamlessly. Does anyone have a clue as to how to get this to work? At the worst case, if someone could list the entire set of operations that are done when installing a WDM kernel streaming driver, Ill make my installl script perform those actions and get rid of the INF file altogether. The basic structure is identical to TESTCAP, so you can see that one's INF file for details. Thanks in advance Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79453
  - 14
    - Serial port number in device manager Hi! I'm working with a driver for virtual serial port and can't make the device manager display the correct port number. It seems to simply increment them, i.e. the first time I create a port it displays e.g. 3, the second time 4 and so on. How can I make the device manager understand which port number is correct? I've set the following: \Device\COMn \DosDevices\COMn (symbolic link) ..\DeviceMap\SERIALCOMM\COMn Thanks in advance, Nille Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79452
  - 15
    - Any news on USB autorun (xp)? Hey all, I've been poking all over the net trying to see what the progress is on getting USB drives to autorun/play. The closest thing anyone has come up with is the funky U3 hack of tricking XP with a phony CD drive work-around that leaves you with /two/ extra letters. Interesting work-around, but a hack, never-the-less. This autorun at least gets your program on the popup window, but it's still not autorun. [autorun] action=Launch Portable Start Menu icon=PStart.ico open=PStart.exe label=Monkey Man! includeRuntimeComponents=True Any progress on getting USB drives to autorun/play? Any word from MS? -b Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79450
  - 16
    - Help! How to get the DriverObject of PCI Let me clarify the problem I am facing with now. We have a PCIe controller which includes a PCI bridge device and another PCI device. Our driver only controls the PCI device while the PCI bridge device is handled by Windows PCI driver. Now due to a hardware problem, I need to access PCI configuration space of the bridge device. Well, by following the rules of WDM, I wrote a PCI bus filter driver to filter pci.sys. This filter driver sends IRP_MJ_PNP/IRP_MN_QUERY_INTERFACE with InterfaceType GUID_PCI_BUS_INTERFACE_STANDARD to the PCI driver to get the PCI_BUS_INTERFACE_STANDARD. And then our original driver communicates with this filter driver and uses PCI_BUS_INTERFACE_STANDARD.ReadConfig/WriteConfig to enumerate and access any PCI devices on the system. This works fine. However, I am looking for a simpler solution (two driver solution is not convinient for user and I don't know if a PCI bus filter driver is elligible for WHQL or not). First of all, I thought I could use IoGetDeviceObjectPointer to get the DO of PCI driver if it is created with a name, unfortunately it looks like the DO is created without a name. Now I thought if I could get the DriverObject of the PCI driver, then I should be able find the PCI DO by walking through the DriverObject->DeviceObject list. I know this list holds the PCI FDO and PDOs for all enumerated PCI devices, however I can tell which is the FDO I needed by checking DeviceObject->Flags. Well, then now I am looking for a documented (or at least stable among different Windows versions) solution to find the DriverObject of PCI. By playing with WinDbg, it looks like I can find the PCI driver by walking through DeviceObject->DeviceObjectExtension->AttachedTo starts with my own DeviceObject. Unfortunately DeviceObjectExtension->AttachedTo is not a documented field. WinDbg command `!drvobj pci` to get the DriverObject of PCI driver. So does WinDbg command `!object \driver\pci`. I don't know how WinDbg implements these commands. Maybe it checks \driver in the object directory and matches with the name, However, it looks like Windows doesn't publish the DDIs needed to query objects under \driver directory. And also I don't know how to interprete the queried data. The OSR tool DeviceTree now is shipped with Windows DDK. I believe it uses undocument fields or APIs. If MS can accept that as its public tool, Could MS make it published to simplify our developers' life? As you see, I am running out of solution now. Maybe I missed something, maybe I am wrong on something, Could anybody here help me to step out of it? Thanks a lot! Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79449
  - 17
    - private data in unidrv Hello! I just want to understand DDK unidrv example. As far as I've read unidrv can include two plug-ins. So,I try OEMDEV structure to send some data from one to another. But when I add some more fields in OEMDEV stucture (I mean besides OEM_DMEXTRAHEADER and dwDriverData, which are happened in DDK example), the data stopped to translate between the plug-ins. Is OEMDEV structure of a fix size, or my private data should be described anywhere more? To fill private members I use the following: LONG APIENTRY OEMDocUICallBack(PCPSUICBPARAM pCallbackParam) { ...... PCBUSERDATA pUserData = (PCBUSERDATA) pCallbackParam->UserData; POEMDEV pOEMDev = (POEMDEV)pUserData->pOEMUIParam->pOEMDM; ........ pOEMDev->dwDriverData1 = 1; pOEMDev->dwDriverData2 = 2; pOEMDev->dwDriverData3 = 3; ....... } Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79447
  - 18
    - AddPrinterDriver I want to install printer through programatically For this i have used AddPrinterDriver function. But i am getting an error saying "1804-The specified data type is invalid. ERROR_INVALID_DATATYPE" I have used DRIVER_INFO_3 structure. The problem is comming while calling the AddprinterDriver function. I am adding the piece of this code Dinfo3.cVersion = 3 ; Dinfo3.pEnvironment = "Windows NT x86" ; Dinfo3.pName = "I TEQ" ; Dinfo3.pDriverPath ="c:\\windows\\system32\\spool\\drivers\\w32x86\\3\\UNIDRV.DLL" ; Dinfo3.pDataFile ="c:\\windows\\system32\\spool\\drivers\\w32x86\\3\\ITEQ.GPD" ; Dinfo3.pConfigFile ="c:\\windows\\system32\\spool\\drivers\\w32x86\\3\\UNIDRVUI.DLL ; Dinfo3.pHelpFile = "c:\\windows\\system32\\spool\\drivers\\w32x86\\3\\UNIDRV.HLP" ; Dinfo3.pDependentFiles ="UNIRES.DLL\0ITEQ.DLL\0ITEQ.GPD\0STDNAMES.Gpd\0\0" ; Dinfo3.pMonitorName = NULL ; Dinfo3.pDefaultDataType = "EMF" ; BOOL RetValue = AddPrinterDriver(NULL,3,(LPBYTE)&Dinfo3); Can anybody help me out of this problem. -Shankar B. Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79444
  - 19
    - Drivers for Toshiba Satellite Pro 6100 Hi, I have installed win 2k pro on my laptop : toshiba satellite pro 6100. On the device manager, i find conflict on "other devices" section. A yellow question mark is near 1- Base System Device 2- Ethernet Controller 3- Multimedia Audio Controller 4- PCI Device 5- TOSHIBA wireless~_LAN_CARD 6- unknown device i don't have the original CD. i looked in toshiba website and used google but could not find the drivers any help please? many thanks Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79437
  - 20
    - NdisQueryBuffer Performance I am working on an NDIS driver and want to look at the Layer 2, IP and TCP headers. I wanted to know how expensive of an operation NdisQueryBuffer is for the buffers that contain the headers. Thanks. Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79435
  - 21
    - Why no DEV_BROADCAST_HDR in WM_DEVICECHANGE ? I am trying to catch up a USB driver insertion/removal with WM_DEVICECHANGE. I called RegisterDeviceNotification() with GUID_DEVINTERFACE_DISK or GUID_DEVINTERFACE_VOLUME, but in WM_DEVICECHANGE message, I can see only DBT_DEVNODES_CHANGED. The parameters like wParam = DBT_DEVICEARRIVAL, and/or lParam= DEV_BROACAST_HDR are never seen. Any suggestion ? My data structure when calling RegisterDeviceNotification() looks like: ZeroMemory(pNotificationFilter, nSize); pNotificationFilter->dbcc_size = nSize; pNotificationFilter->dbcc_devicetype = DBT_DEVTYP_DEVICEINTERFACE; pNotificationFilter->dbcc_classguid = InterfaceClassGuid; memcpy(pNotificationFilter->dbcc_name,deviceName,sizeof(deviceName)); *hDevNotify = RegisterDeviceNotification(hWnd,pNotificationFilter,DEVICE_NOTIFY_WINDOW_HANDLE); Thanks -Gary Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79434
  - 22
    - PsSetLoadImageNotifyRoutine question on win2kSP4 Hello, In my driver i have registered PsSetLoadImageNotifyRoutine. On windows 2000 Sp4 i only see *.exe and ntdll.dll loaded, as compare to xp where i see every image file loaded. IS it something expected ??. The reason I am asking because i was expecting on kernel32.dll notification and act accordingly after kernel32.dll is mapped into process. Thanks Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79419
  - 23
    - MmMapViewInSystemSpace on XP 64 and S2K3 64 bugs But I have a problem with mapping in system address space via PVOID BaseAddress = NULL; SIZE_T RegionLength = length; NTSTATUS rc = MmMapViewInSystemSpace ((PVOID)hSection, &BaseAddress, &RegionLength); Sometime one file mapped on already used and referenced BaseAddress for another file. It's only happened on x64. Anybody have this problem? Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79414
  - 24
    - your help urgently required.......... hello all, i am anil, Mr.Bruno said that alex posted the solution to my situation in some other group, I am accessing this group using OUTLOOK EXpress. but i am unable trace that group. can u help to specify that group. i will seach over there for solution. I nearly worked on it for 4 hrs. But i didnot get it. i am really sorry, because i am again creating new thread here. by specifying problem once again. i have got some opensource project. that is related to USB sniffing. tha t is snoopypro. Actalluy this is developed under Microsoft visual c++. i am getting the folloeing problems at following code. can u update me please, this code atuclly in winddk.h static __inline struct _KPCR * KeGetCurrentKPCR(VOID) { ULONG Value; __asm__ __volatile__ ("movl %%fs:0x18, %0\n\t" : "=r" (Value) : /* no inputs */ ); return (struct _KPCR *) Value; } errors: 1.c:\Documents and Settings\Anil_Desktop\Desktop\SnoopyProSrc\SnoopyPro\wind dk.h(124): error C2065: '__asm__' : undeclared identifier 2.c:\Documents and Settings\Anil_Desktop\Desktop\SnoopyProSrc\SnoopyPro\wind dk.h(124): error C2146: syntax error : missing ';' before identifier '__vola tile__' 3.c:\Documents and Settings\ANil_Desktop\Desktop\SnoopyProSrc\SnoopyPro\wind dk.h(125): error C2143: syntax error : missing ')' before ':' 4.c:\Documents and Settings\ANUPAMA_Desktop\Desktop\SnoopyProSrc\SnoopyPro\w inddk.h(127): error C2059: syntax error : ')' 5.c:\Documents and Settings\ANUPAMA_Desktop\Desktop\SnoopyP roSrc\SnoopyPro\w inddk.h(124): error C3861: '__volatile__': identifier not found, even with argument-dependent lookup and some other warnings. if anybody know correct solution , i require it urgently. i think here we need inputs. If you identify those inputs, we can solve(i think). plz suggest me.. i will wait for your replies. thank u all for spending time to read it... thank u, Anil kumar Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79412
  - 25
    - how to get the video stream format (filter driver) Hi, I have a filter driver for a webcam and I need to do some processing on the video stream before sending it to the user app. The problem that I am facing is that the camera supports multiple video formats (both multiple resolutions and video encoding: RGB and YUV). In the DDK documentation (the "KS Data Formats and Data Ranges" article) it is said that the properties KSPROPERTY_PIN_DATARANGES, KSPROPERTY_PIN_CONSTRAINEDDATARANGES, KSPROPERTY_PIN_PROPOSEDATAFORMAT and KSPROPERTY_PIN_DATAINTERSECTION are used to querry the data format of the pin, and KSPROPERTY_CONNECTION_DATAFORMAT to set the data format (or some property of KSPROPERTY_Connection) I have looked through all the property IOCTLs that my filter receives and I did not find any KSPROPERTY_CONNECTION_DATAFORMAT ; the only IOCTLs from this class that pass through my driver are KSPROPERTY_CONNECTION_STATE and KSPROPERTY_CONNECTION_ALLOCATORFRAMING. Does anyone know how is the stream format set at this level (basic IOCTLs parsing), or at least a point where to look from? This should be closely related to the way AVStream or USBCAMD uses to set the format. BTW, as far as I could guess, it seems that the current format is the one negociated by the last KSPROPERTY_PIN_DATAINTERSECTION, but i did not find any information that it also sets the format, not only querry for a supported one. Can anyone confirm this? Also, I believe that the function driver below me I believe is a USBCAMD 2.0 minidriver. Thanks, Bogdan. Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79409
- Next
  - 1
    - What error does it mean I met this assert in W2K3 checked build with my IM driver and verifier turned on: *** Assertion failed: MiLockedCommit > 0 *** Source File: ..\mi.h, line 3415 What error does it mean ? Peter Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79403
  - 2
    - Please, someone at Microsoft could answare this issue!? Question is marked at the end of this mail!! I'm not able to avoid LanMan to process a request on \IPC$ I've my own redirector and I actually success the prefix claim to MUP: Return SUCCESS to CreateSrvCall Return SUCCESS to CreateVNetRoot My redirector is the first in NP order list. After this I should be the owner of \myserver\IPC$ so I espect to really fail an OPEN on such netroot without LANMAN to take the control. Application try to open \myserver\IPC$\srvsvc and \myserver\IPC$\wkssvc and I've tried to return: OBJECT_NOT_FOUND NOT_SUPPORTED NOT_IMPLEMENTED INVALID_HANDEL and so on... after every successfull call to my prefix claim functions LANMAN always try to open \myserver\IPC$ over the net, it holds win32 application for a long... LANMAN must get a timeout before returning... QUESTION ** I'd like to know if RDBSS addresses the \IPC$ OPEN to LANMAN without asking MUP to get the right provider!!!! Any of the STATUS code returned to CreateVNetRoot or Create successfull avoid LANMAN to keep control of the request!!!! Is it my fault or a RDBSS strange behaviour!? Thanks to all Ale Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79395
  - 3
    - KMDF1.1 with Windows 2000 support KMDF 1.1 with Windows 2000 support is available for download at http://www.microsoft.com/whdc/driver/wdf/KMDF_pkg.mspx. -- -Eliyas This posting is provided "AS IS" with no warranties, and confers no rights. http://www.microsoft.com/whdc/driver/kernel/KB-drv.mspx Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79394
  - 4
    - USB Composite Device, Serial + Mass Storage, could not install ser Hi All, My current project need to implement a composite device which includes Serial and Mass Storage. and without any special driver on windows. I have made the Serial and Mass Storage work as a single function. To install serial, I just create inf file then windows can find and setup it correctly. But, after I create a new composite device, I can still install the serial port on windows by a new inf file but there is question mark (?) on my virtual serial port and it could not be opened. I check my firmware and all the USB request has been responsed. on windows XP, some things, during the copy file of usbser.sys, windows xp will reboot. The mass storage device can work without any problem. Are there anybody here who know if it is possible to install a composite device (CDC+Mass Storage) without and customer driver? (inf file is OK) Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79391
  - 5
    - Can't install Windows Driver Foundation on Windows Xp 64 bit. Error code: 2203 See subject. Bye, Skybuck. Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79390
  - 6
    - Regarding SCSI miniport driver in 64-bit Hi All, I am trying to port a SCSI miniport driver for a virtual HBA that is working without any problem in 32-bit to 64-bit. When I install the HBA (by plugging it to a virtual bus which is created using toaster sample), the DriverEntry() is called followed by FindAdapter(). after that the driver is unloaded and the device manager gives the CM error code 10. (failed to start). The miniport driver is not using any resources. Following are the values I use for the HW_INITIALIZATION_DATA and PORT_CONFIGURATION_INFORMATION: HW_INITIALIZATION_DATA: 1. HwInterrupt(), HwDmaStarted() and HwAdapterState() are not used (set to NULL). Other functions are used. 2. NeedPhysicalAddresses is set to FALSE. 3. MapBuffers is set to TRUE. 4. NumberOfAccessRanges is set to 0. 5. TaggedQueuing is set to FALSE 6. MultipleRequestsPerLu is set to FALSE. 7. AutoRequestSense is set to FALSE. 8. ReceiveEvent is set to FALSE. 9. AdapterInterfaceType is set to Isa. One thing I noticed is that the ScsiPortInitialize() returns STATUS_SUCCESS.. In PORT_CONFIGURATION_INFORMATION, 1. InterruptMode is set to LevelSensitive. 2. CachesData is set to FALSE. 3. MaximumNumberOfTargets = 6 (just for testing) 4. BufferAccessScsiPortControlled = TRUE 5. ScatterGather = FALSE 6. MapBuffers = TRUE 7. NeedPhysicalAddresses = FALSE 8. TaggedQueuing = FALSE 9. MultipleRequestsPerLu = FALSE 10. AutoRequestSense = FALSE 11. Master = FALSE. Actually After FindAdapter(), the IRP_MN_REMOVE_DEVICE is called for the adapter in the bus driver which unloads the driver. So, there should be some problem in FindAdapter() that causes this. But same code works fine in 32-bit without any problem. When I examined the event log, I got following error: Driver detected an internal error in its data structures for \Device\Scsi\MyDrv1. For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp. 0000: 00 00 00 00 01 00 56 00 ......V. 0008: 00 00 00 00 04 00 04 c0 .......Ã? 0010: df fa ff ff bb 00 00 c0 Ã?ÃºÃ¿Ã¿Â»..Ã? 0018: 00 00 00 00 00 00 00 00 ........ 0020: 00 00 00 00 00 00 00 00 ........ Thanks, Killme. Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79383
  - 7
    - Devcon sample weirdness Does anyone know why the following: _NT_TARGET_VERSION=$(_NT_TARGET_VERSION_WIN2K) is necessary in the Vista WDK Devcon sources file for device enable/disable functionality to work? I don't really have time to dig through the devnode right now to figure this out. I am hoping someone from MSFT might be able to give us a heads up on this? Thanks. Bill M. Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79381
  - 8
    - HDD SMART Status Requests Does anyone know how to turn off the default behaviour in XP, Win 2000 and 2003 which insists on sending SMART Status requests to IDE devices periodically. I am trying to write software to measure HDD operation times and these automatic requests are interrupting the program flow. Is this SMART checking configurable in the registry or something? Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79380
  - 9
    - Device manager Hi! I've been given a legacy "software-only" driver that provides virtual serial ports. The problem is that the driver doesn't support pnp, so the serial ports doesn't show up in the device manager (only FDOs there, right?). What would be the simplest way to make the device manager aware of the ports? I've tried making the driver pnp-compatible, but that is starting to seem like a dead-end. First, I made a bus driver that hosted the virtual serial ports. Worked fine except for the fact that windows refused to automatically select the drivers, so when the user installed a serial port (which can be done at any time) he/she needed to locate the drivers. Not good. I used SetupCopyOEMInfo to make windows aware of the inf-file - my theory is that windows won't automatically select a driver that is not signed. I also tried doing a user-mode installation of the port, using the code from Walter Oney's Fastinst program. That works fine as long as only one port is installed. If several ports are installed, each new installation causes all existing ports to be reinstalled by UpdateDriverForPlugAndPlayDevices. Again, not good. Any ideas? Thanks in advance, Nille Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79370
  - 10
    - Limit on data that can be sent through ExtEscape I am working on a unidrv printer driver. In order to provide some custom functionality, I need to pass some data to the graphics dll (spooler context) from the ui dll (application context). I am using ExtEscape to do this. This works fine except that I do not receive the full data in some cases. I believe this happens when the data size is too large. Is there any limit on the size of the data that can be transferred through ExtEscape. Thanks in advance. Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79369
  - 11
    - DIFx bus driver install Hi, I would like to use DIFx to install a bus driver, this bus driver is very simular to the one in toaster WDK sample. Everything seems fine but looking at the log file: DIFXAPP: INFO: No matching devices found in INF "C:\WINDOWS\system32\DRVSTORE\...drv.inf" on the Machine. DIFXAPP: INFO: No drivers installed. No devices found that match driver(s) contained in 'C:\WINDOWS\system32\DRVSTORE\...drv.inf". DIFXAPP: SUCCESS:Installation completed with code 0xE000020B. DIFXAPP: INFO: RETURN: DriverPackageInstallW (0xE000020B) DIFXAPP: The device(s) for which the driver is being installed has not been plugged into the computer. It turns out that the bus driver is pre-installed but not started. If I want to install it manually, I need to indicate that the hardware is already connected in the "Add/Remove hardware" feature. I guess that here is the issue. How can I tell DIFx to start the bus driver after it installed it ? I did not find any sample for that, the toaster one only focuses at a toaster device DIFx not the busd driver one. Is this a DIFx limitation ? Thank you for helping me, Jean Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79365
  - 12
    - Bsod - filter manager Hi, I'm trying to determine why my machine crash during startup, and by doing an analyze on the stack dump, it seems to point to filter manager. However, I think there might be something else, but I'm unsure how to proceed. Surely it must be some dodgy drivers being loaded but how do I find out? Listed below is the information. cheers, Kenny KERNEL_MODE_EXCEPTION_NOT_HANDLED_M (1000008e) This is a very common bugcheck. Usually the exception address pinpoints the driver/function that caused the problem. Always note this address as well as the link date of the driver/image that contains this address. Some common problems are exception code 0x80000003. This means a hard coded breakpoint or assertion was hit, but this system was booted /NODEBUG. This is not supposed to happen as developers should never have hardcoded breakpoints in retail code, but ... If this happens, make sure a debugger gets connected, and the system is booted /DEBUG. This will let us see why this breakpoint is happening. Arguments: Arg1: c0000005, The exception code that was not handled Arg2: 806ff94f, The address that the exception occurred at Arg3: f52daaf4, Trap Frame Arg4: 00000000 EXCEPTION_CODE: (NTSTATUS) 0xc0000005 - The instruction at "0x%08lx" referenced memory at "0x%08lx". The memory could not be "%s". FAULTING_IP: hal!HalAllocateCommonBuffer+3f 806ff94f ?? ??? TRAP_FRAME: f52daaf4 -- (.trap fffffffff52daaf4) Unable to read trap frame at f52daaf4 CUSTOMER_CRASH_COUNT: 1 DEFAULT_BUCKET_ID: DRIVER_FAULT BUGCHECK_STR: 0x8E LAST_CONTROL_TRANSFER: from baef8c06 to 806ff94f STACK_TEXT: f52dab78 baef8c06 867180f0 857c1c20 00000000 hal!HalAllocateCommonBuffer+0x3f f52dabdc baf091a7 857c1c20 867a63e0 00000000 fltMgr!FltpGetStreamListCtrl+0x5a f52dabf8 baef5ac9 857c1c20 867a63e0 857c1c20 fltMgr!FltpCleanupStreamListCtrlForFileObjectClose+0x17 f52dac14 baef5ffb f52dac2c 8642c460 8695c280 fltMgr!FltpPassThrough+0x93 f52dac44 804e13d9 8642c460 882ace00 806ff428 fltMgr!FltpDispatch+0xf3 f52dac44 867a63c8 8642c460 882ace00 806ff428 nt!KiTrap0B+0x10f WARNING: Frame IP not in any known module. Following frames may be wrong. 00000000 00000000 00000000 00000000 00000000 0x867a63c8 STACK_COMMAND: kb FOLLOWUP_IP: fltMgr!FltpGetStreamListCtrl+5a baef8c06 ?? ??? FAULTING_SOURCE_CODE: SYMBOL_STACK_INDEX: 1 FOLLOWUP_NAME: MachineOwner SYMBOL_NAME: fltMgr!FltpGetStreamListCtrl+5a MODULE_NAME: fltMgr IMAGE_NAME: fltMgr.sys DEBUG_FLR_IMAGE_TIMESTAMP: 41107bad FAILURE_BUCKET_ID: 0x8E_fltMgr!FltpGetStreamListCtrl+5a BUCKET_ID: 0x8E_fltMgr!FltpGetStreamListCtrl+5a Followup: MachineOwner --------- Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79362
  - 13
    - NDIS problem indicating from MiniportSendPackets Hi I have written an NDIS IM driver that do some filtering of packets. This is basically what is does: * In MiniportSendPackets - save off a copy of a TCP SYN packet and notify a user mode application about this using a shared event object. The original packet is discarded (not send further down with NdisSend). * The user mode application tells the driver how to rewrite the saved SYN packet which in the same call allocates a new NDIS_PACKET and indicate this packet up with NdisMIndicateReceivePacket. * The packet arrives to a user mode application listening on the destination address of the rewritten packet. This accepts the connection which sends a SYN/ACK that I catch in MiniportSendPackets. I here make a copy of the SYN/ACK and modifies the copy once again by changing source and destination ip/port, recalculate the checksum and then indicate this copy up with NdisMIndicateReceivePacket. The problem is that this copy never seems to get through the tcpip.sys layer - that is not going to the socket matching the source/destination ip/port of the packet. I can't understand why this works in the first step but not in the second. The only difference between the two are that in the first case it is (it might be) running at a different IRQL but this should not make any difference. Can there be a problem that the SYN/ACK copy is being indicated before the original SYN/ACK has been completed? This doesn't sound very likely since they have nothing with eachother to do (atleast not that the upper layer knows of). I have tried both copying the original OOB data and not doing this but without any difference in behaviour. The same with setting packet flags. I don't use NDIS_SET_ORIGINAL_PACKET since the two packets should not be linked together. I use the same handle for the MiniportAdapterHandle for the two indications and I allocate the packets from the same pool (same PoolHandle). I do the same procedure for packets coming off the wire which works great so I only have problems with the packets that are being looped back without going out on the LAN. The difference between the two are that in the case where packets should be looped back I need to indicate all packets from the MiniportSendPackets and in the "normal" case from the LAN I only indicate them from ProtocolReceive/ProtocolReceivePacket and filter outgoing packets in MiniportSendPackets with NdisSend. I've tried to look at the TDI layer for the packet but it doesn't show up there so it must be lost in between the upper NDIS and lower TDI which I guess is the tcpip.sys driver. Is there a way of debugging or logging what is happening at the upper NDIS layer or in tcpip.sys? I can't use ethereal since this doesn't work well with IM drivers and communication on the same machine. Everything would probably be solved if I could guarantee that a packet which has the same destination MAC-address as the sending machine would always be looped back by the underlying driver so that I could process the packet normally in the ProtocolReceive/ProtocolReceivePacket instead. Is there a way of forcing an underlying layer to do loopback? I also must be able to tell that the packet is originating from the local machine (to avoid spoofing). Any suggestions on why I can't get this to work or some tips on how to figure out what is causing the problem? Magnus Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79355
  - 14
    - Efficient serial port access We have a kernel-mode DLL that needs to detect a signal event (CTS) taking place in a serial port. The 10us wide signal comes at exact 1000ms intervals from the accurate external clock. By comparing the performance counter values at the moment of event detection, we have noticed that there is a fluctuating delay in detecting the CTS event. The 0.25ms delay stepwise accumulates for 3-7 successive events (up to even 1,6 ms) and then resets by itself. This kind of delay is too much for our time critical solution and invalidates the benefits of the accurate GPS clock used. Below, I have included the code for listening the serial port event. Is anyone able to provide a faster way of detecting the CTS pulse? The listening takes place in its own thread and its priority is the highest possible. Does the port's read timeout have an effect on detection delay? Thanks Ali //***************************************************************************** // WaitForCTS // Waits for the CTS event to take place in a serial port. // aDeviceObject - Port device object. // Returns the status code. //***************************************************************************** NTSTATUS WaitForCTS( IN PDEVICE_OBJECT aDeviceObject ) { NTSTATUS status; PIRP irp; KEVENT event; IO_STATUS_BLOCK io_status; ULONG events_occurred = SERIAL_EV_CTS; KeInitializeEvent( &event, NotificationEvent, FALSE ); // set up an IRP for waiting events irp = IoBuildDeviceIoControlRequest( IOCTL_SERIAL_WAIT_ON_MASK, aDeviceObject, NULL, 0, &events_occurred, sizeof( ULONG ), FALSE, &event, &io_status ); // send the IRP to the driver status = IoCallDriver( aDeviceObject, irp ); if ( status == STATUS_PENDING ) { // wait for the completion status = KeWaitForSingleObject( &event, Executive, KernelMode, FALSE, NULL ); } if ( status != STATUS_SUCCESS ) { if ( status != STATUS_TIMEOUT ) { DbgPrint( "Error %#X\n", status ); } else { KdPrint(( "Timeout\n" )); } return status; } return io_status.Status; } Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79350
  - 15
    - How to use WriteFile() to write WIA properties to the diveice? Hi, everyone When I develop a Scanner, I have a problem how to use WriteFile() to write WIA properties to the hardware. For example, before getting a image from a scanner, we must write device-specific settings (such as WIA_IPS_XRES, WIA_IPS_YRES, WIA_IPS_BRIGHTNESS etc.) to the hardware. The scanner how to know which data is WIA_IPS_XRES or WIA_IPS_BRIGHTNESS. Thanks! Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79343
  - 16
    - How to register Vendor-defined IOCTL fail message to Eventlog? Hello Everyone: In my device driver , I defined some vendor-defined IOCTL by CTL_CODE macro, IOCTL value between 2048~4095. When the IOCTL return customize error code, Windows Eventlog will add a record of ErrorMessage, but the message is "IOCTL UNKNOW .........". How could I register some more useful message to replase it ?? Thank you for your attention.! Haikun Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79337
  - 17
    - Obtain Serial Number As I can obtain the Serial Number of a Pen Drive. I must obtain from a unit specifies example: E:\ Tanks! Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79333
  - 18
    - Does Win98 SmartCard Service support USB reader hot plug-in ? Hello Everyone: My Win98 Reader driver works well if reader plugged before Win98 boot. But if I plugged reader after Win98 boot, seems pc/sc manager couldn't talk to driver correctly. From driver debug, I found all driver PnP enumerate IRP worked well, but driver couldn't get smartcard service request. I wonder if pc/sc manager support USB reader hot plug-in in Win98? Thank a lot for your attention!! Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79332
  - 19
    - How to safely write EA to a file at file system filter driver level Hi folks, I have a file system filter driver that constantly monitor file changes. When I detect a file change, I would write EA (Extended Attribute) to that file. There doesn't seem to exist a documented kernel API to do it. So by reverse engineering, I created a way to do it. By reverse engineering, I mean I interceptd the ZwSetEaFile call, and looked at how irp was built, and mimic the irp operation. This sometimes caused problem. The symptom IS MFT corruption error. I figure may be I run into some synchronization issue because of not holding the internal lock used by ntfs. Does anybody have any suggestion on how to fix this problem? Thanks. Daniel Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79329
  - 20
    - Processor Affinity and friends Dear All Let's take a dual-processor machine for example: If I set the processor affinity in IoConnectInterrupt to only one CPU, then the ISR and all code synchronised with it will run on the specified CPU. Could somebody confirm if this is right as well, please : - If an application talks to my driver directly (no filter drivers involved), and that application is running on both CPUs, then all code in the driver executed by the calling thread of the application will run on the same CPU as the application thread was running on. I.e. there is no CPU switching performed when calling into the kernel. - What CPU are DPCs executed on? As interrupt affinity does not apply to DPCs, I would think that the DPC could be executed on any processor. - What CPU are Power Management and PnP Request carried out on? As per my understanding, this is executed on any of the two processors, depending on the calling (Operating System) thread. Following on from this: - You can tie entire applications to run on only one processor by using 'Imagecfg.exe'. Is there a way of forcing all driver code to be executed on only one processor? I am sure the answer is NO, but I want to be mega-sure! :-) Thanks Christian Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79321
  - 21
    - Access to Performance Counters in 'System Object' in Kernel Mode d Is there any other way to access Performance Counters in Kernel Mode driver? One way is to use WMI application and push the data to Kernel Mode Driver thru DeviceIoControl. Thanks, Best, Srini Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79318
  - 22
    - Enabling NDIS Trace I am trying to get NDIS Tracing to work with little success. I am working with the free build of Windows XP SP2 and a checked version of NDIS.sys. I use the symbol server, and lm shows my ndis symbols are loaded: 0: kd> lm start end module name 804d7000 80701000 nt (pdb symbols) c:\symbols\ntkrnlmp.pdb\7DE39A3E89DA4B378B95A09FA3A6398C2\ntkrnlmp.pdb f744c000 f7478a80 NDIS (pdb symbols) c:\symbols\ndis.pdb\42ED3DC0817A4246B157736BBAF668742\ndis.pdb However trying to get/set the debug level through the debugger I get the following error: 0: kd> !dbglevel NDISKD: error - could not access ndis!ndisDebugLevel - check symbols for ndis.sys I guessed the symbol server downloads symbols for the free version of NDIS.sys, so I went to my symbols folder and replaced the ndis.pdb file with one taken from a host debugging a checked build (indeed the files differ) by issuing: copy C:\CheckedSymbols\ndis.pdb\DD85689443D34D1D95DF405FEBFFA5AA1\ndis.pdb C:\symbols\ndis.pdb\42ED3DC0817A4246B157736BBAF668742 but then I got messages about mismatched symbols. Can anyone tell me what am I doing wrong? Thanks Hillel Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79317
  - 23
    - PsSetLoadImageNotifyRoutine Dear all, At what IRQLs can my PsSetLoadImageNotifyRoutine notification routine be called at? The IFS documentation isn't clear on this point but for PsSetCreateThreadNotifyRoutine and PsSetCreateProcessNotifyRoutine callbacks it clearly states that the notification routine will be called at PASSIVE_LEVEL. Many thanks in advance, Carly Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79315
  - 24
    - Child display driver ... adding PnP crashes! Hello I want to implement a WDM Child Display Driver and request I2C interface from display driver which I am going to use to send the DDC/CI data to Monitor. Intially, just to test the I2C interface, I added the I2C routines in the AddDevice() function and was able to get a pointer for the I2CInterface. But there might be a problem with this, since I was never able to call i2copen() function after this.( The system crashes ). Later I tried to modify the driver, and add the required dispatch routines. But as soon as i add the PnP dispatch routine the system crashes!! My DriverEntry() looks like this --------------------------------------------------------------------------------------------------------------------------------- DriverObject->MajorFunction[IRP_MJ_CREATE] = I2C_WDMDispatch; DriverObject->MajorFunction[IRP_MJ_CLOSE] = I2C_WDMDispatch; DriverObject->MajorFunction[IRP_MJ_DEVICE_CONTROL] = I2C_WDMDispatch; DriverObject->DriverUnload = I2C_WDMDriverUnload; DriverObject->MajorFunction[IRP_MJ_PNP] = I2C_WDMDispatchPnp; //PnP routine DriverObject->MajorFunction[IRP_MJ_POWER] = I2C_WDMDispatchPower; //Power Mangement DriverObject->MajorFunction[IRP_MJ_SYSTEM_CONTROL] = I2C_WDMDispatchSystemControl; DriverObject->DriverExtension->AddDevice = I2C_WDMAddDevice; // AddDevice ------------------------------------------------------------------------------------------------------------------------------ and the IoCreateDevice function has been called with the following parameters -------------------------------------------------------------------------------------------------------------------------- IoCreateDevice ( DriverObject, // our driver object sizeof (FDO_DATA), // device object extension size NULL, // FDOs do not have names FILE_DEVICE_VIDEO, // Video device FILE_DEVICE_SECURE_OPEN, // FALSE, &deviceObject); // The device object created ---------------------------------------------------------------------------------------------------------------------------- Can anyone give me some tips as to why adding the PnP routine forces the system to crash Please let me know if you want some further information Thanks and regards Abhishek Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79310
  - 25
    - Why !deadlock command is not functional ? I have this bug: DRIVER_VERIFIER_DETECTED_VIOLATION (c4) ... Parameter 1 = 0x1000 .. 0x1020 - deadlock verifier error codes. Typically the code is 0x1001 (deadlock detected) and you can issue a '!deadlock' KD command to get more information. Arguments: Arg1: 00001001, subclass of driver violation. Arg2: 864fedd8 Arg3: 84fdacf0 Arg4: 00000000 But when I am trying: 0: kd> !deadlock issue: 00001001 00000000 864FEDD8 FFFFFFFF No deadlock detected ----- What can be reason that in my debugger is not functional this command ? Peter Tag: Why doesn't mirror driver catch BitBlt sometimes? Tag: 79307

my mirror driver can catch BitBlt call. But I noticed that it's failed
to catch BitBlt sometimes immediately after I change the screen
resolution. I want the mirror driver be able to capture the whole
screen at first by using
BitBlt(hdc2,0,0,SCREEN_WIDTH,SCREEN_HEIGHT,hdc1,0,0,SRCCOPY)). I found
that BitBlt was executed successfully, But mirror driver didn't capture
it. why? BTW, if I right click the task bar and select show the
Desktop, the mirror driver may capture the whole screen always.

Does anyone help me out? any comment?

Top