How to force re-enumeration of all devices

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
    - Direct connection problem in WinCE Hi, I am trying to establish direct connection from our COM port in WinCE to the host which is a XP pc. I made the necessary setup on XP and on WinCE when I make a new instance of Direct Connection for my COM port to operate with XP with the same settings like baud rate flow control etc... given for XP, it says " Connecting to HOST " and then "You have been disconnected from the Remote Computer...Please verify the baud rate setting and retry the connection". Here I am not settingup any active sync connection. Can any one tell me how to fix this... Thanks in advance Srinivas Tag: How to force re-enumeration of all devices Tag: 101156
  - 2
    - Windows Update driver available ? Hi, we have succesfully passed WHQL test for a video capture device. From ou r submission, it is indicated that Windows Update driver is available. However, when I plug camera on a computer where camera has not been plugged before, Vista system searchs on Windows Update but cannot find the driver. What must be perform to allow installation from Windows Update after the submission processed is Ok? Damien Tag: How to force re-enumeration of all devices Tag: 101154
  - 3
    - Issues about USB composite device driver I want to develop an USB composite device driver. The device has one configuration and 6 interfaces in it. According desription of DDK(2003 sp1, 3790.1830) documents, Windows system(XP and later) had supplied an USB common class generic parent driver, that is usbccgp.sys. This driver can be loaded when our device pluged in and create 6 PDO for each interface. So we can install driver for each interface. But I have found some problems about this description. 1. OS can't find the usbccgp.sys driver and install it automatically for our device. I had searched the c: partition(my system disk) but can't find the usbccgp.sys file. My Windows system is XP sp2. Why and where can I get the usbccgp.sys file? 2. I write a inf file and reference usbhub.sys to install driver for our device. Yes! usbhub.sys can work! It enumrates 6 devices from our device for each interface. Then I install drivers for each device. I can install driver for interface number 0(matched ID is USB\VID_xxxx&PID_xxxx&MI_00) and the driver work well. But when I try to install driver for other interface, it failed! The failed position is when I try to select configuration for USB device with USBD_CreateConfigurationRequestEx and submit it to host driver. Laterly, I found some useful information from DDK document. It said that host driver can't support selecting configuration in composite USB device. But the problem is that where can I get the interface information and pipe information(ep address and pipe handle etc.) if I can't get those information from selecting configuration? Please help me, thank you very much! Abei.liu 12/05/2007 Tag: How to force re-enumeration of all devices Tag: 101153
  - 4
    - WHQL certification of laptop - USB issue Hello I'm stuck with strange behaviour of the USB subsystem preventing requirements for WHQL certification of a laptop with an embedded USB device. First things first. Here's an interesting test: I'm experiencing a strange phenomenon with USB devices that are surprise-removed during standby. I've seen it happening with USB hubs, USB sticks, USB hard drives and USB 3G devices. It happens in Win2000 and WinXP. These are the steps taken: 1. Plug in one of the mentioned USB devices. 2. In Bootvis, initiate standby test. 3. During standby, unplug the USB device -> surprise-removal. 4. Resume. 5. Wait for Bootvis to finish. In Bootvis you can see that the device driver of the unplugged USB device eats an extra +/- 3 seconds of resume time (compared with the resume time in a situation where you do not unplug the device). The exact device driver depends on the USB device; I've noticed the 3 extra seconds with usbhub.sys, usbcgpp.sys, usbstor.sys, and gt72ubus.sys (3th party 3G driver). When you unplug multiple USB device during standby, the system eats 3 seconds for each unplugged device. I found out that during resume from standby, the USB subsystem eats 3 seconds for each unplugged device during standby. Technically: When a USB function driver sends down the set D0 power request for an unplugged device, some lower-level code happily sleeps for 3 seconds and completes the request. At this point, the USB function driver doesn't know of the surprise-removal yet; that event happens later. Can anyone tell me why Win2000/XP behaves like this? So in my case: The laptop has a built-in embedded USB function, a MiniPCIe card. The laptop de-powers the card during standby, hence a surprise-removal happens and resume from standby takes an extra 3 seconds. As a result, the total resume time for this laptop takes more than 5 seconds. The laptop needs to be WHQL-certified (logo) for WinXP. Because of the long resume time, the laptop doesn't meet the requirements. Pretty annoying. Any suggestions? Tag: How to force re-enumeration of all devices Tag: 101152
  - 5
    - STATUS_UNSUCCESSFUL after calling WdfDeviceStopIdle Hi experts, I'm facing a problem with my KMDF USB driver when coming back from idle (D2 power state) My USB driver receives no IRPs (or WDFREQUESTS for that matter). Data passes through it only via direct function calls from an NDIS miniport on a different stack. I've set the idle timeout to be 1 millisecond and the idle setting are by default "off", i.e. I call WdfDeviceStopIdle as soon as the miniport connects to USB driver Whenever I want to go to low power state, the driver calls WdfDeviceResumeIdle and the driver goes to D2. All work fine until this point. Whenever I want to restore my communication with the device I call WdfDeviceStopIdle. In the context of the EvtDeviceSelfManagedIoRestart (after we're in D0) I send the request. The problem is that the requrest fails with status 0xC0000001. The USBD status of the failure is 0x0. Same failure occurs for both read and write request. Same failure occurs even if the request is sent not from the context of the EvtDeviceSelfManagedIoRestart context but rather from the context of a dedicated workitem which is spawned from the EvtDeviceSelfManagedIoRestart context. I used a USB sniffer to sense what is going on when the error occurs; I see that there are no Start of Frames (SOFs) at all! When I increase the idle timeout to about 250 milliseconds the problem disappears. I have no explanation for this. I can supply the KMDF log of the failure if anyone wishes to see. I've looked at it but saw no explanation for the failure. What could be the reasons for this error status? My guess is that the USB host controller hasn't resumed from the selective suspend state although the framework has already sent me an IoRestart event. Please help. Thanks, Yoav. Tag: How to force re-enumeration of all devices Tag: 101147
  - 6
    - USB 2.0 ISO Allways at Beginning of SOF? Using a high-bandwidth ISO endpoint ( 3- 1024 packets), and only one USB ISO device in system. Question: Since the system allocates a maximum of 80% bandwidth the ISO endpoint, is it guaranteed that the ISO xfer (IN token) will ALWAYS occur at the beginning of the microframe(SOF)? So if there is a control xfer or something else in the same microframe, at SOF the ISO xfer is serviced first, then the remaining 20% at the end of the microframe for the control transfer? I didn't see anything in the USB 2.0 spec. Thanks -decoder Tag: How to force re-enumeration of all devices Tag: 101139
  - 7
    - IoCompletion and free the driver-allocated IRP By attaching buffers for isochronous transfer with 1394 kmdf driver: the t1394_IsochAttachBuffers function passes the IRP to next-lower driver with IoSetCompletionRoutine( newIrp, t1394_IsochAttachCompletionRoutine, IsochDetachData, TRUE, TRUE, TRUE ); IoCallDriver(deviceExtension->StackDeviceObject, newIrp); Then the driver calls t1394_IsochAttachCompletionRoutine. The IoCompletionRoutine checks with 'if (!NT_SUCCESS(Irp->IoStatus.Status))', ; ---------------------------------------------------------------------------------------------------------------------------------- Irp->IoStatus.Status is 0x0, I think, it means STATUS_SUCCESS, so !NT_SUCCESS(Irp->IoStatus.Status) == FALSE. ---------------------------------------------------------------------------------------------------------------------------------- jumps to end of this IoCompletionRoutine and returns 'STATUS_MORE_PROCESSING_REQUIRED'. The t1394_IsochAttachBuffers function returns STATUS_PENDING. So the nStatus will be stay by STATUS_PENDING forever and the application freezes by calling GetOverlappedResult(hDevice, &overLapped, &dwBytesRet, TRUE) function. So far I have understood... In the msdn article about 'IoCompletion', I saw these line : If an IoCompletion routine returns STATUS_MORE_PROCESSING_REQUIRED, the lower driver's call to IoCompleteRequest immediately returns. In this case, a higher-level driver will have to call IoCompleteRequest to complete the IRP. I found the WdfRequestCompleteWithInformation function in my source code, but this will be call just for the case !NT_SUCCESS(Irp->IoStatus.Status) == TRUE (I think, this is the case for invalid parameters). So, shouldn't be there also a WdfRequestComplete function for !NT_SUCCESS(Irp->IoStatus.Status) == FALSE ? thanks in advance min Tag: How to force re-enumeration of all devices Tag: 101137
  - 8
    - symbol path problems 1) I have symbols on network drive like u:\localsym. I can access the path using windows explorer(win explorer first asks password and validates password). After windows explorer access the network drive, I open windbg and windbg gives warniong "inaccessible symbol path u:\localsym". Spelling is correct. I dont know what is going wrong here. 2) If one of the symbol path is wrong, will windbg try to check for symbols from other paths in symbol path or will it abondon all together? Tag: How to force re-enumeration of all devices Tag: 101132
  - 9
    - ReadEX Can someone explain to me how ReadEX with overlap works on wdf type driver if works? On the wdm was simply the read function of driver return PENDING status then process IO. On the wdf there is not return status from EvtIoRead function and WdfRequestMarkCancelable does not causing system to return PENDING status to ReadEX. Tag: How to force re-enumeration of all devices Tag: 101129
  - 10
    - A simple file access monitor hello, I need to write a application that can monitor file/share access and the users who did the operation. I know they are things out there that get close to it like the FileSystemWatcher() funcion in .Net that almost fits my needs except that it doesn't know who did the changes. Also I don't want to user syslog because it is not designed for this and also not really fast. Right now all I could find on the net that could help me along was to implement a minifilter myself. Is it possible to find out the user who performed an operation on a file or a share with a minifilter driver? Best Regard and thank you for any help. V. Tag: How to force re-enumeration of all devices Tag: 101118
  - 11
    - Occasional performance issue with EngModifySurface We have an application that draw an animation into a layered window as fast as possible. I post here because the problem is related to win32k and EngModifySurface. Please tell me if another group is a better place for this post. On some hardware the animation is always fast. On others hardware, sometimes the animation is slow, sometimes the animation is fast. This occur immediately after the window is created. When the window is closed and reopened (without closing the process ), it can again be slow or fast, but this never switch after the initialisation of the window. I profiled the application in the two cases using Kernrate. The logs tell us that there is more time spend in the function EngModifySurface in win32k when the application is slow and that something is slowing down EngModifySurface around the address 0xbf8be470 and 0xbf8be4c0. Does anyone know what happen at this address, or a way to know it and how this could be fixed so that the animation is always fast ? Here is the simplified log of Kernrate in the slow case first and in the fast case in second. All the measures were done on the same computer. Thank you very much for attention. Emmanuel Caradec * KERNRATE LOG : SLOW CASE Time 7343 hits, 25000 events per hit -------- Module Hits msec %Total Events/Sec intelppm 4099 10000 55 % 10247500 win32k 2649 10000 36 % 6622500 <--- This is very high ntkrnlpa 344 10000 4 % 860000 hal 138 10000 1 % 345000 ialmdev5 64 10000 0 % 160000 ===> Processing Zoomed Module win32k.sys... ----- Zoomed module win32k.sys (Bucket size = 16 bytes, Rounding Down) -------- Percentage in the following table is based on the Total Hits for this Zoom Module Time 2649 hits, 25000 events per hit -------- Module Hits msec %Total Events/Sec EngModifySurface 2568 10000 97 % 6420000 <--- This is very high EngStretchBlt 24 10000 0 % 60000 EngDeleteSurface 9 10000 0 % 22500 EngCreateBitmap 7 10000 0 % 17500 EngModifySurface+0x8cc[0xbf8be470] , 48711, 1212 < Disasm (-16 bytes | bucket content | +16 bytes): <--- many hits at 0xbf8be470 EngModifySurface+0x8dc[0xbf8be480] , 48712, 40 < Disasm (-16 bytes | bucket content | +16 bytes): EngModifySurface+0x8ec[0xbf8be490] , 48713, 20 < Disasm (-16 bytes | bucket content | +16 bytes): EngModifySurface+0x8fc[0xbf8be4a0] , 48714, 22 < Disasm (-16 bytes | bucket content | +16 bytes): EngModifySurface+0x90c[0xbf8be4b0] , 48715, 24 < Disasm (-16 bytes | bucket content | +16 bytes): EngModifySurface+0x91c[0xbf8be4c0] , 48716, 1270 < Disasm (-16 bytes | bucket content | +16 bytes): <--- many hits at 0xbf8be4c0 * KERNRATE LOG : FAST CASE OutputResults: KernelModuleCount = 133 Percentage in the following table is based on the Total Hits for the Kernel Time 5846 hits, 25000 events per hit -------- Module Hits msec %Total Events/Sec intelppm 3379 10000 57 % 8447500 ntkrnlpa 803 10000 13 % 2007500 ialmdev5 789 10000 13 % 1972500 win32k 443 10000 7 % 1107500 <--- This is normal hal 342 10000 5 % 855000 USBPORT 17 10000 0 % 42500 ===> Processing Zoomed Module win32k.sys... ----- Zoomed module win32k.sys (Bucket size = 16 bytes, Rounding Down) ------- Percentage in the following table is based on the Total Hits for this Zoom Mode Time 443 hits, 25000 events per hit -------- Module Hits msec %Total Events/Sec EngModifySurface 174 10000 46 % 435000 <--- This is normal EngStretchBlt 100 10000 26 % 250000 EngDeleteSurface 31 10000 8 % 77500 FONTOBJ_pxoGetXform 11 10000 2 % 27500 EngSetLastError 10 10000 2 % 25000 EngUnmapFontFileFD 8 10000 2 % 20000 EngFreeUserMem 8 10000 2 % 20000 EngModifySurface+0x8cc[0xbf8be470] , 48711, 14 < Disasm (-16 bytes | bucket content | +16 bytes): <--- few hits at 0xbf8be470 EngModifySurface+0x8dc[0xbf8be480] , 48712, 7 < Disasm (-16 bytes | bucket content | +16 bytes): EngModifySurface+0x8ec[0xbf8be490] , 48713, 12 < Disasm (-16 bytes | bucket content | +16 bytes): EngModifySurface+0x8fc[0xbf8be4a0] , 48714, 19 < Disasm (-16 bytes | bucket content | +16 bytes): EngModifySurface+0x90c[0xbf8be4b0] , 48715, 41 < Disasm (-16 bytes | bucket content | +16 bytes): EngModifySurface+0x91c[0xbf8be4c0] , 48716, 17 < Disasm (-16 bytes | bucket content | +16 bytes): <--- few hits at 0xbf8be4c0 Tag: How to force re-enumeration of all devices Tag: 101114
  - 12
    - KMDF: Use of KeSetImportanceDpc()? I'm writing a KMDF driver for a network device with very small paket buffers. I transfer data from the device in DPC. Latency between interrupt and DPC call is critical, because the DPC has to be called within typical 250-500 microseconds after the device raises its interrupt, else pakets are lost. To solve this, I'd like to experiment with "KeSetImportanceDpc()". Questions: - I think I need to use WdfDpcWdmGetDpc(), but how do I access the WDFDPC-handle of a DPC, which was declared indirectly with .... WDF_INTERRUPT_CONFIG_INIT(&interruptConfig, MY_ISR, MY_DPC); status = WdfInterruptCreate(pDeviceContext->Device, &interruptConfig, WDF_NO_OBJECT_ATTRIBUTES, &pDeviceContext->Interrupt); .... ? - Where should I call KeSetImportanceDpc() ? In the ISR directly before WdfInterruptQueueDpcForIsr()? Thanks, J. Hoppe, PEAK System Technik Tag: How to force re-enumeration of all devices Tag: 101113
  - 13
    - How to record audio is playing on my app with virtual audio device ? Hello everyone, i want to recording audio is playing on my app. I get virtual audio driver from WDM and i installed Microsoft Virtual Audio Device (Simple) WDM . 1. I want to copy sound from my app to this virtual audio device. 2. I want to copy sound from this device to real recording device. Anyone can help me how i can do step1 and/or 2 with Microsoft Virtual Audio Device (Simple) WDM ? Must i implement copyfrom and copyto in BaseDMA.cpp ? Many thank for any suggest. Tag: How to force re-enumeration of all devices Tag: 101112
  - 14
    - video child device in Windows Vista Hi !! I want to create a video child device using video miniport in Windows vista, so I 'm writting non pnp kernel mode driver which creates a FDO and attaches the fdo to PDO (e.g \\Device\\Video3) using IoAttachDeviceToDeviceStack. Now How can i create child device for this video device ? Thanks in Advance Mangesh Tag: How to force re-enumeration of all devices Tag: 101111
  - 15
    - msvcr80.dll and msvcp80.dll used in in print driver I am developing a print driver which contains a dll which depends on msvcp80.dll and msvcr80.dll. How do I add these dependents to my INF such that they are managed properly? Do I need to include the manifest file somehow? For the heck of it, I tried treating them like any other file in my INF and was able to install the driver, but the printer failed to install, and I received a "Resolve Partial Assembly failed for Microsoft.VC80.CRT. Reference error message: The referenced assembly is not installed on your system." in the event viewer. The dependency walker looks fine when examining files in the destination user mode directory. I've had no luck finding an answer in the MSDN or WDK docs. Tag: How to force re-enumeration of all devices Tag: 101107
  - 16
    - slow boot windows xp Hello, For a couple of mounts, I'v a slow boot of windows xp on my acer laptop 9412. First, I thought a virus or rootkit but the tools I tried don't find anything of important (S&D, AVG, Bitdefender, SDoctor). Starting in save mode works better but also slow. Bootvis isn't helping me very well. It only shows often 100% disk use during boot which I knew already by the HD led. Watching processes with process explorer from sysinternals, I see a lot of interrupts during boot; around 45% CPU use during boot. Using process monitor, I see a long file time in the file summary for c:\$ConvertToNonresident (around 45% of total file time). I suppose there is an NTFS problem but defragmentation doesn't help. I hope this newsgroup has the knowledge what is the problem or what causes it. What is happening ? How do I solve this slow boot problem ? In the stack summary, I see some <unknown>. I suppose that's normal ? Tag: How to force re-enumeration of all devices Tag: 101102
  - 17
    - NtShutdownSystem Vista compatibility Hello everyone, I am using NtShutdowSystem in my native application which is working great with Windows Xp and in case of Vista, it restarts system properly but afterwards gives a "Windows Recovery Screen" (hope this is familiar) of which if we choose "Start Windows Normally", everything works fine. The problem is why I am getting such a screen in Vista and not in Xp? How can this be fixed? Adding to above, autochk.exe in Vista does a restart without any problem at all, so there must be some way for sure. Thanks in advance. Tag: How to force re-enumeration of all devices Tag: 101100
  - 18
    - Win32 API for Tape Backup Hello all, I am looking at creating a tape backup routing using Win32 API (CreateFile, CreateTapePartition,WriteFile,ReadFile,...). And those function descriptions just confuse me. One particular function: 'PrepareTape'. I just don't know when I need to call this function. And how does I know the tape is at ' \\.\Tape0'? regards, Peggy Tag: How to force re-enumeration of all devices Tag: 101098
  - 19
    - CreateFile fail to open device I am working on a mouse filter driver which has two device objects. One is the conventional device object shown in DDK, the other one is a named device object transfering some data out. The question is I can find two objects in namespace viewer but createfile() can not open the secound named object. Code in driver .h #define NT_DEVICE_NAME_STRING L"\\Device\\MIOCTL" #define DOS_DEVICE_NAME_STRING L"\\DosDevices\\MTest" // ==Symbolic_name_string .c RtlInitUnicodeString(&ntDeviceName, NT_DEVICE_NAME_STRING); RtlInitUnicodeString(&symbolicLinkName, DOS_DEVICE_NAME_STRING); status = IoCreateDevice(DeviceObject->DriverObject, sizeof(CONTROL_DEVICE_EXTENSION), &ntDeviceName, FILE_DEVICE_UNKNOWN, FILE_DEVICE_SECURE_OPEN, FALSE, &ControlDeviceObject); status = IoCreateSymbolicLink( &symbolicLinkName, &ntDeviceName ); Code in app (hDevice = CreateFile( "\\\\.\\MTest", GENERIC_READ | GENERIC_WRITE, 0, NULL, OPEN_EXISTING, FILE_ATTRIBUTE_NORMAL, NULL) The test application returns system error code #2 "can not find file" . Who can tell me how to open this device in filter driver? Is there some utilities I can view the dos devices? Thanks in advance, Tag: How to force re-enumeration of all devices Tag: 101097
  - 20
    - Driver Problem with Canon LBP2900 Printer Hi Folks, Unfortunately, I encounter a problem with my brand new laser printer Canon LBP2900. It is connected to my computer (Windows Vista) via an USB cable and works properly as long as I am offline. But once I go online, data transfer from the computer to the printer will take several minutes. This phenomenon seems not to depend on the size of the file to be printed. Pages are not printed stepwise (with time interruptions between different pages) but in one single go once printing is initiated. I suspected the firewall (Norton) to be responsible for this delay. But even if I turned it off, delayed printing did persist. I have already uninstalled the printer driver and reinstalled when I was online with no effect on the slowed data transfer to the printer. During data transfer, the program I want to print from (Word, Adobe Acrobat etc.) seems to be pretty busy (task manager will temporarily show 'Not Responding'). When I want to print a PDF, the green status bar will rest at zero percent for ages. Then (after several minutes) it will abruptly turn to 100%. I would be very pleased if somebody could give me some advice here. Thank you so much. Regards Ulf Tag: How to force re-enumeration of all devices Tag: 101094
  - 21
    - some help please Hi, I am a newbie in the driver development and i am planning to do some research about a project. What i need to do is to create a virtual printer. When i print to that printer i need to show a form for some settings, after i hit print on this form i need to save the printjob in some kind of filetype. That's all i need to do. Can someone point me in the right direction what is the best way to do this? Many Thanks! FC Tag: How to force re-enumeration of all devices Tag: 101091
  - 22
    - SpinLock in DPC and IOCTL Hi, I use a spin lock to synch my IOCTLs (which can be called from several user mode clients), and my DPC. How will the system behave in the next scenario: IOCTL xxx has been called, and starts to run in PASSIVE_LEVEL and aquires spin lock MyLock. During its excecution, an interrupt has been triggered by the hardware, and a DPC starts to run in DISPATCH_LEVEL, aquiring MyLock. Will it waits to MyLock to be released (which means that DPC waits for IOCTL to be completed) or will it run ignoring the lock (because its in higher IRQL)? Thanks, Avi. Tag: How to force re-enumeration of all devices Tag: 101089
  - 23
    - how can I kernel debug NDIS IM driver "passthru" example in VMWare?! After I install the "passthru" in vmware succesfully, then connect to vmware using windbg, and set some breakpoint on "PtReceivePacket" or MP* functions, it simply doesn't stop there. I use bridged mode in vmware for the NIC. I can surf the internet in the vmware machine, does this mean that the packet doesn't go through "passthru"?? Thanks for any reply. Tag: How to force re-enumeration of all devices Tag: 101081
  - 24
    - Running DTM in VMWare I tried to run the NDISTest for my CoNDIS WAN virtual virtual adapter. Since the current NDISTest is boundaled with DTM, I need to install the DTM controller and studio in a server in oder to run the NDISTest. What I did is as follows. 1. Install DTM controller and studio in a server runnning a VM. 2. Install my CoNDIS Virtul WAN and DTM client in Windows XP running in a another VM. In the above configuration, I can create the machine pool to monitor the job without any problem. But when I tried to create a submission, it always fail in a few steps after running the Wizzard. Does anybody have any experiece to run the DTM on a Vmware? If I cannot run the DTM in the VMware, can I just run the NDISTest in the VMWare? then how? Tag: How to force re-enumeration of all devices Tag: 101079
  - 25
    - EvtWmiInstanceExecuteMethod OutBufferSize problem Hello, I wrote a driver which supports a WMI ExecuteMethod. I wrote a Vbscript program to test the method but the OutBufferSize in the EvtWmiInstanceExecuteMethod is aways 5 which is not big enough for the return data. The method takes inputs and returns parameters including and array of data. Why is the OutBufferSize always 5 ? Is there away to tell the Swbemobject what size I want the Output buffer to be ? Usually the input buffer is the output buffer from the EvtWmiInstanceExecuteMethod point of view but I don't know how SWBemojbect come up with a OutBuffersSize of 5. What am I doing wrong ? vbscript and mof below thanks in advance --- vbscript On Error Resume Next Dim Locator Dim Service Dim Collection Dim InstancePaths() Dim InstanceCount Dim CurrentInstanceIndex Set S = CreateObject("WbemScripting.SWbemNamedValueSet") Set Locator = CreateObject("WbemScripting.SWbemLocator") Set Service = Locator.ConnectServer(, "root\wmi") Service.Security_.ImpersonationLevel=3 Set Collection = Service.InstancesOf ("Microsoft_IPMI") InstanceCount = 0 Err.Clear for each Instance in Collection if Err.Number = 0 Then InstanceCount = InstanceCount + 1 ReDim Preserve InstancePaths(InstanceCount) Set ObjectPath = Instance.Path_ InstancePaths(InstanceCount) = ObjectPath.Path 'MsgBox Instance.Path_ Else MsgBox Err.Description End If next 'Instance Err.Clear if InstanceCount = 0 Then MsgBox "No instances available for this class" Else CurrentInstanceIndex = 1 Set Instance = Service.Get(InstancePaths(InstanceCount )) Call RequestResponse End if Sub RequestResponse ' Obtain an InParameters object specific to ' the IpmiInterface.RequestResponse method. Set objInParam = Instance.Methods_("RequestResponse").inParameters 'Wscript.Echo objInParam.Properties_.Count 'Wscript.Echo objInParam.Properties_.item("RequestData").CIMType 'Wscript.Echo objInParam.Properties_.item("RequestData").Name ' Add the input parameters. ' ipmi get device id objInParam.Properties_.item("NetworkFunction") = 6 objInParam.Properties_.item("Lun") = 0 objInParam.Properties_.item("ResponderAddress") = 32 objInParam.Properties_.item("Command") = 1 objInParam.Properties_.item("RequestDataSize") = 0 objInParam.Properties_.item("RequestData") = Array ' Call RequestResponse Set objOutParams = Instance.ExecMethod_("RequestResponse", objInParam) Wscript.Echo objOutParams.ReturnValue If Error = 0 Then Wscript.Echo "RequestResponse.Ccode =" _ & objOutParams.CompletionCode Else Wscript.Echo "RequestResponse error " End If End Sub --- mof sniplet of method RequestResponse [Implemented, WmiMethodId(1), Description ("" ) : amended ] void RequestResponse( [in, Description("Network Function for this IPMI request.") : amended ] uint8 NetworkFunction, [in, Description("Logical Unit number to send IPMI request to") : amended ] uint8 Lun, [in, Description("Responder address to send IPMI request to") : amended ] uint8 ResponderAddress, [in, Description("IPMI request command") : amended ] uint8 Command, [in, Description("Bytes for IPMI request data") : amended ] uint32 RequestDataSize, [in, WmiSizeIs("RequestDataSize"), Description("IPMI request data") : amended ] uint8 RequestData[], [out, Description("Completion code for status of request response") : amended, ValueMap { "0","192","193","194","195", "196","197","198","199","200", "201","202","203","204","205", "206","207","208","209","210", "211","212","213","255", "1..126","128..190",".." }, Values { "Command Completed Normally", "Node Busy", "Invalid Command", "Command invalid for given LUN", "Timeout while processing command", "Out of space", "Reservation Canceled or Invalid Reservation ID", "Request data truncated", "Request data length invalid", "Request data field length limit exceeded", "Parameter out of range", "Cannot return number of requested data bytes", "Requested Sensor, data, or record not present", "Invalid data field in Request", "Command illegal for specified sensor or record type", "Command response could not be provided", "Cannot execute duplicated request", "Command response could not be provided. SDR Repository in update mode", "Command response could not be provided. Device in firmware update mode", "Command response could not be provided. BMC initialization or initialization agent in progress", "Destination unavailable", "Cannot execute command. Insufficient privilege level", "Cannot execute command. Command, or request parameter(s), not supported in present state", "Unspecified error", "Device specific (OEM) completion codes", "Standard command-specific codes", "reserved" } ] uint8 CompletionCode, [out, Description("Bytes for IPMI response data. Completion code is present in response data size.") : amended ] uint32 ResponseDataSize, [out, WmiSizeIs("ResponseDataSize"), Description("IPMI response data. Completion code is present in response data.") : amended ] uint8 ResponseData[] ); -------------------------------- Tag: How to force re-enumeration of all devices Tag: 101078
- Next
  - 1
    - Writing a USB device driver for a trackball I have a mouse and a trackball on my pc. I would like the mouse to act as normal mouse that controls the cursor. I would like the trackball to act as a separate device which does not act like a mouse. Right now, when I plug both in, I get two mice. The trackball uses HID-compliant mouse driver in the device manager. I was hoping to write a new driver and install it instead of the HID- compliant driver. I could then use the new driver in my application to simply read delta x and y from the trackball. Could anyone give me some ideas on this? Is there any similar source code that I could modify? I open to other methods too. I examined using Raw input but ran into so roadblocks. Tag: How to force re-enumeration of all devices Tag: 101075
  - 2
    - no audio device hi everybody, I had to format a pc,"hp compaq dx2000 mt", I've downloaded all the drivers from the hp's website for this pc's model, but the audio doesn't work. I've the Operative System Win Xp pro, and when I go in "managment device" , it says "the device works properly". Instead when I go in "Sounds and Audio Devices" in "control panel" it says "No audio device". I've already tried to find alternative drivers in "Windows update" uninstalling previously the hp's audio driver, but it was unable to find anything. Can someone help me? Thank you very much Alex Tag: How to force re-enumeration of all devices Tag: 101071
  - 3
    - Printer UI Plugin Hi, after reading a lot of documentation about UI and rendering plug in , I would like to know if I well understand. with rendering plug in, I can intercept the data flow sent to te printer through the print job , for example. and with UI plug in, I can add property page to the print document or print device, and store setting data into the registry. Now, I think that I can only communicate with the printer through the rendering plug in, and so , I wonder how to send data directly to my printer from my UI plug in without to execute a print job. I think I don't understand all the process of the printing. If anyone can help me, I really appreciate. and sorry for my poor english. thank you Tag: How to force re-enumeration of all devices Tag: 101068
  - 4
    - Commtimeout values Hi all, I need to set timeout values for readfile and writefile, the timeout for readfile is 300milliseconds and for writefile is 20seconds. I read the MSDN documentation, but I m not clear to specify the timeout valuesCan any one help on this. COMMTIMEOUTS cto; cto.ReadIntervalTimeout = ?; cto.ReadTotalTimeoutMultiplier=? cto.ReadTotalTimeoutConstant =?; cto. WriteTotalTimeoutMultiplier=?; cto. WriteTotalTimeoutConstant =?; Regards, Hari Tag: How to force re-enumeration of all devices Tag: 101061
  - 5
    - PrintMonitor GPD File I am working on a virtual image print driver using a print monitor. Once installed of the orientation is set in the control panel (CP->printing preferences) it is honored just fine, however if the orientation is set by the application at print time, it's not working. I believe this might be something I need to change in the gpd file, but I'm not sure, any pointers on where to start looking? Peter Tag: How to force re-enumeration of all devices Tag: 101053
  - 6
    - vista & 1394 virtual driver I got 1394 virtual driver working on Windows XP SP2. I tried to use same binary on Windows Vista. I tried to create virtual 1394 driver by sending a request to \\.\\1394BUS0 device. Strangely I get an error "Access denied" when I try to do CreateFile(). I dont understand what happened here. I am using admin account. hDevice = OpenDevice(NULL, pCurrentBusName, FALSE); HANDLE OpenDevice( HANDLE hWnd, PSTR szDeviceName, BOOL bOverLapped ) { HANDLE hDevice; DWORD dwError = SFB_SUCCESS; if (bOverLapped) { hDevice = CreateFile( szDeviceName, GENERIC_WRITE | GENERIC_READ, FILE_SHARE_WRITE | FILE_SHARE_READ, NULL, OPEN_EXISTING, FILE_FLAG_OVERLAPPED, NULL ); } else { hDevice = CreateFile( szDeviceName, GENERIC_WRITE | GENERIC_READ, FILE_SHARE_WRITE | FILE_SHARE_READ, NULL, OPEN_EXISTING, 0, NULL ); } if (hDevice == INVALID_HANDLE_VALUE) { dwError = GetLastError(); SFB_Print(" Failed to open device %s, error %ld \r\n", szDeviceName, dwError); } return(hDevice); } // OpenDevice Thanks, Raj Tag: How to force re-enumeration of all devices Tag: 101051
  - 7
    - What does it take to get PCI bus master privileges in Vista? I have a pair of NDIS drivers that sit on top of a bus driver that controls a PCI device. In Vista32 one NDIS driver that sits on top of our bus driver calls NdisMSetMiniportAttributes with the attributes flag set to NDIS_MINIPORT_ATTRIBUTES_BUS_MASTER. I can allocate shared memory. In Vista64, the same driver, the same source code, the same call to NdisMSetMiniportAttributes, with the same attributes flag set to NDIS_MINIPORT_ATTRIBUTES_BUS_MASTER, and I CAN NOT allocate shared memory. What other things does this NDIS driver have to do to join the 64 bit bus master club? Thanks -Ron- Tag: How to force re-enumeration of all devices Tag: 101043
  - 8
    - IoCompletion The IoCompletion function is defined : NTSTATUS IoCompletion( IN PDEVICE_OBJECT DeviceObject, IN PIRP Irp, IN PVOID Context ); What does IoCompletion do with the third parameter 'IN PVOID Context' ? In MSDN, I see 'Caller-supplied pointer to driver-specific context information, Context information must be stored in nonpaged memory', but I couldn't find out more detail... Can somebody explain it? Thanks in advance, min Tag: How to force re-enumeration of all devices Tag: 101032
  - 9
    - Problem With Device Class GUID Hi All, I have developed a USB-Serial driver which installs two driver. One under USB Controller and other under ports. I have GUID explorer application which lists all the hardware devices. I could see my USB device under the USB device class guid in the lists. But i could not see the COM device under Port device class GUID (GUID_DEVCLASS_PORTS). Can any one let me know if i am missing any initialization settings in serial port installation? Any help is appreciated. Thanks and Regards, Sushma. Tag: How to force re-enumeration of all devices Tag: 101030
  - 10
    - Vitsa driver load reload problem I know I've seen commants on this before but Google is just swamped with stuff relating to 'Vista' and 'driver'. I have a little test utility to exercise a driver I'm working on. As a convienence, it contains buttons to spawn devcon commands to install/update/restart my driver. The update button (via the .inf) causes the latest build of my driver to be copied to system32\drivers and then fires up the new code. All this works great on XP. On Vista, either the copy takes place and it gets overwritten by the previous version or the copy never takes place. The automatic driver restore monster seems to be in the game. Is there any magic that can be applied from the command line (i.e. devcon) to make Vista behave? The driver is in a system defined class and is signed using a test certificate (pending receipt of real certificate) Thanks, Mickey. Tag: How to force re-enumeration of all devices Tag: 101028
  - 11
    - NdisAllocateBuffer help??? Hi, I would like to ask a questiobn about NdisAllocateBuffer. Whenever VirtualAddress and length are passed to this method. Is Address's containt copied to buffer or just address is mapped to buffer(which is returned by api). Please sovle this small query. -- With regards thanks Anand Choubey Tag: How to force re-enumeration of all devices Tag: 101025
  - 12
    - Send User Handle to Driver Hello I want to send a user handle to my driver. Try to use METHOD_BUFFERED io_ctl, but seems that the IoDeviceControl callback function is not called, when init nInBufferSize ( DeviceIoControl 4th parameter). What is the correct way to do that? Thanks Tamar Tag: How to force re-enumeration of all devices Tag: 101023
  - 13
    - NDIS5 wrapper BSOD I have a filter driver, which works well until a legacy driver (USB- based network card) gets initialized. The system crashes right after my filter driver initiates packet sending (not realy, the packet belongs to my driver). This is my stack: nt!DbgBreakPointWithStatus+0x4 nt!KeBugCheckEx+0xc01 nt!Kei386EoiHelper+0x28ec usbhub+0x1d53 ndis!NdisFreeNetBufferListContext+0x460 ndis!NdisMWanSendComplete+0x1ce ndis!NdisMSendComplete+0x4d3 FA101ND5+0x768 ndis!NdisMSendComplete+0x1833 ndis!NdisMSendComplete+0x16d8 ndis!NdisMFreeNetBufferSGList+0xfb ndis!NdisFSendNetBufferLists+0x18 filter_driver!OriginalNdisSend+0x89 filter_driver!SendPacket+0x292 ..... Is there anything specific I should do with the packet in Vista filter driver before submitting it to NdisFSendNetBufferLists? If packets are simply relayed or if there is no legacy drivers installed, my driver works flawlessly. Thanks in advance, Gene Tag: How to force re-enumeration of all devices Tag: 101022
  - 14
    - how to use pvOut in DOCUMENTEVENT_CREATEDCPRE Hi, The documentation given in MSDN for pvOut variable for function IPrintOemUI2:DocumentEvent with iEsc DOCUMENTEVENT_CREATEDCPRE is as follows: Pointer to a driver-supplied DEVMODE structure, which GDI uses instead of the one supplied by the CreateDC caller. (If NULL, GDI uses the caller-supplied structure.) I am implementing my own ui plug in. In this plug in i am trying to change the values of some of the public devmode member variables. I tried assigning the new values to members variables of pvOut. Even though the change is reflected in iEsc DOCUMENTEVENT_CREATEDCPOST, the change is not reflected in the print job. ie if I change pvOut- >dmDuplex to DMUP_HORIZONTAL from DMUP_SIMPLEX, the print job is still simplex. The above problem is observed in MSWORD and other office applications. It works fine for notepad, adobe reader etc. Can I any one point out the right way to store the devmode settings? Tag: How to force re-enumeration of all devices Tag: 101021
  - 15
    - x64 driver not working with 32bit apps I have ported a 32 bit driver to XP x64 and it works with x64 applications like IExplorer x64. It also prints a test page successfully. However, it will not work with any 32bit applications. When I run Windbg with a 32bit app and envoke print, I don't see any calls to my x64 driver like DrvEnableDriver or DrvEnablePDEV. Any ideas as to what would cause this? Don L. Tag: How to force re-enumeration of all devices Tag: 101010
  - 16
    - Trust webcam wb 5400 Hi All, I am trying to download the Vista drivers for my webcam but the download from Trust site fails always, the file is named 15007_03.exe. Do somebody knows where I can find it, I have found some sites but they aren't Trust sites. Those drivers are right ? Sometimes these sites want to be paid.... Thank you in advance. Tag: How to force re-enumeration of all devices Tag: 101009
  - 17
    - StorPortInitializeDpc & PASSIVE_LEVEL WDK claims: "The StorPortInitializeDpc routine must be called during HBA initialization from within the miniport driver's HwStorPassiveInitializeRoutine routine." Whatâ??s the point to call StorPortInitializeDpc this way? StorPortInitializeDpc is a wrapper over KeInitializeDpc, that can be called at any IRQL TIA Andrew Tag: How to force re-enumeration of all devices Tag: 101006
  - 18
    - upgrading DTM We are currently running DTM from WLK 1.0a and are trying to upgrade to WLK 1.0c. The tests have to be unistalled before DTM can be uninstalled. Add / remove tests generates the error "This action is only valid for products that are currently installed." Any ideas what is going on? Tag: How to force re-enumeration of all devices Tag: 100998
  - 19
    - How to detect programmatically whether someone is using the computer Suppose there is a Windows XP open and running some software and somebody sits in front of PC and begin to use/interact with this software (so creating process is not a clue). Any good idea? I guess the only way to do it is monitorizing keyboard or GUI events. But... is it possible to "read" the queue of events (or something like that) from a kernelmode driver? Keyboard aproximation seems too similar to keylogger and I would prefer other "no so intrusive" solution. Any suggestions are welcome Regards, Juli=E1n Tag: How to force re-enumeration of all devices Tag: 100994
  - 20
    - Switching from USB to Bluetooth host I'm trying to write a driver for a bluetooth device that is not recognized as such by windows. However, it is recognized when it's connected to USB. Maybe it would be possible to use Windows DDK to connect to the device and then change the address so that the reports get received via the bluetooth hub. I have the driver up and running for USB and I only consider bluetooth an option if it doesn't require a rewrite of the driver. Has someone done something similar to this before or can anyone give me some pointers? Best regards, Niels Keetels Tag: How to force re-enumeration of all devices Tag: 100991
  - 21
    - A question about I/O operation. When I call ReadFile in user mode and it's a blocked operation, so when the I/O completes, the ISR will schedule a DPC to signal the completion of I/O to the I/O manager, and the I/O manager then will copy the buffer from the driver to the user mode buffer, but how the thread calling ReadFile become awaken?! Will the I/O manager schedule a APC to the thread to wake up it?? If I said wrong, please feel free to correct me and give the answer here. Thanks really!!! Regards Tom Tag: How to force re-enumeration of all devices Tag: 100989
  - 22
    - DTM Testing and DriverStudio Hello All, We have a driver which is based on WDM model, this driver has been compiled using driverstudio. Now we are planning to get logo for vista under unclasssified cateogry (DTM Test). Is it possible to get logo from driverstudio based compiled driver ? (I asked this question as somewhere it has been mentioned that to get logo testing, driver should be compiled using WDK) -- Tahnks, Nash Tag: How to force re-enumeration of all devices Tag: 100988
  - 23
    - Using Event Object in overlapped I/O operation : GetOverlappedResult is blocked... In 1394api.dll source code (:\WinDDK\6000\src\1394\dll\1394api\isochapi.c) for 1394 WDM driver, there is an overlapped I/O operation for 'Isochronous Attch Buffers'. It creates a manual-reset event object : ------------------------------------------------------------------------ overLapped.hEvent = CreateEvent( NULL, TRUE, FALSE, NULL ); ------------------------------------------------------------------------ with a ResetEvent function : ------------------------------------------------------------------------ ResetEvent(overLapped.hEvent); ------------------------------------------------------------------------ After that, it calls DeviceIoControl function with &overlapped : ------------------------------------------------------------------------ dwRet = DeviceIoControl( hDevice, IOCTL_ISOCH_ATTACH_BUFFERS, pIsochAttachBuffers, ulBufferSize, pIsochAttachBuffers, ulBufferSize, &dwBytesRet, &overLapped); ------------------------------------------------------------------------ It waits for pending operation by calling GetOverlappedResult function with : -------------------------------------------------------------------------------------------------- if (!GetOverlappedResult(hDevice, &overLapped, &dwBytesRet, TRUE)) { //getoverlappedresult failed, lets find out why... dwRet = GetLastError(); TRACE(TL_ERROR, (hWnd, "IsochAttachBuffers: GetOverlappedResult Failed! dwRet = %d\r\n", dwRet)); } --------------------------------------------------------------------------------------------------- Unfortunately the sample application freezes by calling GetOverlappedResult function. As far as I understood, GetOverlappedResult function with 'bWait = TRUE' does not return untill the pending operation has completed. But I can't find the 'SetEvent function' for our manual-reset event object. Shouldn't there a SetEvent function to set the manual-reset event object explicitly to the signaled state? Could these be the cause of application freezing? Or did I miss something? Can somebody explain this? Thanks in advance, min p.s On DebugView, it seems that the pending operation has completed... Tag: How to force re-enumeration of all devices Tag: 100986
  - 24
    - IFS Filter and File Sizes Dear all, I am writing a small filter driver to hone my skills and I am not sure if something I am trying to do will produce guaranteed results! I would like to be able to always know the size of a file (end of file length) based on filtering Irp and Fast IO writes and set informations. This seemed like a valid approach. However, am I right in thinking that if a file is memory mapped, the size could change without getting an Irp or Fast IO call via the cache manager? In which case there would be times that I would not know the size of a file unless I asked the FS/Cache Manager? Many thanks in advance? Carl Tag: How to force re-enumeration of all devices Tag: 100974
  - 25
    - Filtering packets in Wireless Vista Hi All, I am trying to develop a filter driver in Vista to read wireless packets. I am using NdisGetDataBuffer API to read packets, but it does not give me the wireless header. Can anyone guide me as to what shall i do to read wireless header in vista filter driver? Cheers, Wizz Tag: How to force re-enumeration of all devices Tag: 100973

How can I get XP to re-enumerate all hardware devices, similar to what would
happen when installing the O/S, yet without requiring a re-install? Scanning
for hardware changes from Device Manager does not pick up (bind) all devices.

Thanks,
Michael

Top

How to force re-enumeration of all devices by J

J
Thu Dec 06 05:44:32 PST 2007

M> How can I get XP to re-enumerate all hardware devices,
M> similar to what would happen when installing the O/S,
M> yet without requiring a re-install?

<URL:http://homepages.tesco.net./~J.deBoynePollard/FGA/put-down-the-
chocolate-covered-banana.html>

M> Scanning for hardware changes from Device Manager
M> does not pick up (bind) all devices.

The real question, that you should be looking for the answer to, is
why bus enumeration is not discovering the devices.

Top

RE: How to force re-enumeration of all devices by Michael

Michael
Thu Dec 06 07:09:00 PST 2007

Good question! The device in question is an Away Mode device
(http://download.microsoft.com/download/a/f/7/af7777e5-7dcd-4800-8a0a-b18336565f5b/awaymode.doc).
Update Rollup 2 was installed. Later, the BIOS was upgraded. Now, on boot
(or 'scan for hardware changes') Windows finds the away mode device, however,
when it tries to bind to it Windows says it can not find the appropriate
driver. In "Installing and Configuring Away Mode," under "BIOS
Modifications" (p. 4) the document infers the away mode device is enumerated
during setup / install. This appears consistent with the observed behavior.

"J de Boyne Pollard" wrote:

> M> How can I get XP to re-enumerate all hardware devices,
> M> similar to what would happen when installing the O/S,
> M> yet without requiring a re-install?
>
> <URL:http://homepages.tesco.net./~J.deBoynePollard/FGA/put-down-the-
> chocolate-covered-banana.html>
>
> M> Scanning for hardware changes from Device Manager
> M> does not pick up (bind) all devices.
>
> The real question, that you should be looking for the answer to, is
> why bus enumeration is not discovering the devices.
>

Top