4 Full speed isochronous->1 High speed isochronous = 100% CPU

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
 - FAT32 File system I need help on the FAT32 file system implementation. Please clarify this doubts. - How does the OS find the size of the new device when the device is formatted? In floppy where the size of the floppy is stored.? - How much space we need to allocate for FAT table in FAT32 file system? Thanks! Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84851
 - 2
 - strange behaviour of tagged pool Hi, I have an NDIS 5.1 miniport. In order to test some mechanism I built I inserted some calls to NdisAllocateMemoryWithTag in the checkforhang callback. This means that every few seconds I allocated a chunk of memory, I did not free it or even read or wrote to it - just allocated it. I used both Poolmon.exe and also !poolusage from WinDbg to monitor pools status and in both cases I got the same results: When the allocated size was 4081 and higher both Poolmon and !poolusage show me that the tagged pool allocations are indeed increased i.e. I saw an evidence to the allocations. But when the allocation size was 4080 or lower neither poolmon or !poolusage showed me any indication to the allocation from the tagged pool. The pool's statistics were unchanged and (I made sure through the debugger that NdisAllocateMemoryWithTag is indeed called). Any idea what is the reason for that? Thanks Miki Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84850
 - 3
 - Sharing interruot between two drivers for the same interface Hi, Can the interrupt be shared between -two drivers on PCI interface, one of the driver being Ndis miniport for that particular interface ? -two drivers on PCMCIA interface, one of the driver being Ndis miniport for that particular interface ? - two driver on USM interface, one of the driver being Ndis miniport for that particular interface ? If so, is status = IoConnectInterrupt( &DeviceExtension->InterruptObject, BcmIrqServiceRoutine, (PVOID) DeviceExtension, NULL, Vector, Irql, Irql, Mode, Shared, Affinity, FALSE ); is the function used in all the non-ndis driver with shared attib and in Ndis the called used is Status = NdisMRegisterInterrupt(&Adapter->Interrupt, AdapterHandle, AiVector, // bus-relative vector number used by the NIC. AiInterrupt, // bus-relative DIRQL for the interrupt TRUE, // MiniportISR function should be called each time the NIC interrupts TRUE, // Other devices on the I/O bus can use this interrupt line InterruptMode); // using LevelSensitive Inerrupts ... even in non-ndis and ndis on USB interface also call the same function to share the interrupts ? Regards Esha Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84848
 - 4
 - multipath on Longhorn Hi Everybody, I'd like to do the FC multipath test on Longhorn. http://www.microsoft.com/WindowsServer2003/technologies/storage/mpio/faq.mspx > Q. Are there any plans to release an "Industry Standard" Device Specific > Module (DSM) for Fibre Channel? > A. Yes. The Microsoft industry-standard DSM for Fibre Channel is planned for > inclusion in Microsoft Windows Server code-named Longhorn. I read this news, and I tried to configure FC multipath, but I could't configure the multipath. Please teach the procedure how to configure multipath by using "Industry Standard" DSM for FC which is included in Longhorn. (or has not been included yet?) Longhorn : beta2 Enterprise Edition 32bit HBA : LP9802 * 2 Thanks Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84846
 - 5
 - KMDF & USB isochronous Hello, I'm currently evaluating KMDF for an USB Driver. Our device is based on Cypress FX2 chip. We stream video trough EP2 configured as IN Isochronous (High Bandwith device - 2*724 bytes per =B5frame). When looking in the documentation (Version 1.1 of the framework), I see that WdfUsbTargetPipeFormatRequestForRead method must be used with a Bulk or Interrupt Pipe. So I assume that an Isochronous Pipe is not a valid pipe for this method. I also saw that the framework has a continuous reader. But my understanding of this reader is that it's purpose is only for intrinsically asynchronous events (like a change in a device state). As this continous reader doesn't have a limitation on which kind of pipe it use, can I use it for my application ? So my question : What is the right method to use a High Bandwidth Isochronous streaming device with KMDF ? Should I create URB's instead ? I must add that while our device streams video. It's not recognized as a steraming device under windows, that is we do not use directshow or other AVStream interface. So the fact that is video that is currently streamed has little importance. Thanks, Fred. Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84840
 - 6
 - About calling User mode functions. Respected Sir, I have a kernel-mode dll, through which I want to call User-Mode Functions. I would like to know asto how can I do so in correct manner ? regards, Sudhanshu Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84830
 - 7
 - accessing ttl field from windows filtering platform? I have an odd but simple requirement that I need to see the ttl field and check its value from incoming IP packets. Currently we do this using a simple .sys filter driver (TDI area). This is not really going to work too well for Vista platforms. I would like to use the Windows Filtering Platform to do this task. Is it even possible to check the TTL field of incoming IP packets using WFP? How would I do this? I have not been able to figure this out from the msdn documentation/library I have found. Thanks for the help. Jay Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84812
 - 8
 - WinXPSP2, Intel Core2Duo, Kernel drivers Hello, I write experimental drivers, and have never worked on SMP systems yet. I need to upgrade to an IBM T60 but I am worried about the dual-core Intel processor engine. I have the following questions: - It seems the dual-core provides 2 pipelines and 2 caches to 2 cores on a single die. This sounds like a SMP system. Will my drivers have to worry about SMP stuff? - Can I configured Windows XP SP2/BIOS to completely disable/ignore one of the cores when I write my drivers? Will there be a major performance impact if I did this? I read that Windows installs a different HAL when it sees an SMP... so it sounds like I cant do this at boot time each time.. it has to be a one-time decision during installation? Basically, I want to know if I can write non-SMP aware drivers in someway on a IBM T60 that has a core2duo processor engine. Thanks! Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84810
 - 9
 - SetupDiCallClassInstaller on Vista I have been using SetupDiCallClassInstaller with DIF_REMOVE to uninstall PCI devices on W2K and XP. This does not appear to work properly on Vista CTP (5472). There is no error returned, i.e, the call to SetupDiCallClassInstaller succeeds, but the device is still installed. Is this a known problem with Vista, or did something change in regards to using the SetupDiXXX API? Thanks, Harry Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84805
 - 10
 - WdfIoTargetStop calls FxUsbPipe::WaitForSentIoToComplete even with WdfIoTargetLeaveSentIoPending? I've set up a completion routine for a bulk endpoint continuous reader using KMDF which reads data into a buffer. If my buffer is about to overflow, I want to stop the reader until the buffer is emptied by my application. I'm at DISPATCH_LEVEL inside the completion routine, so I call WdfIoTargetStop() with argument WdfIoTargetLeaveSentIoPending. According to the docs, this is the only way it's acceptable to call WdfIoTargetStop at IRQL <= DISPATCH_LEVEL. However, this still blue screens for me. When I look at the stack backtrace, it looks like WDF is still trying to wait for some sort of cleanup, which causes the blue screen. You can also see from the trace that I passed 0x00000003 to WdfIoTargetStop() which is the enumerated value for leave-sent-IO-pending. Any ideas? 8047166c f38d1aef 8365e748 8364e690 00000000 Wdf01000!FxUsbPipeContinuousReader::CancelRepeaters+0x31 8047167c f38ca634 8364e690 00000000 00000000 Wdf01000!FxUsbPipe::WaitForSentIoToComplete+0x109 80471694 f38c613d 00000003 83b940d0 8364e690 Wdf01000!FxIoTarget::Stop+0x41 804716b0 f396d61a 8365e800 7c9b1968 00000003 Wdf01000!imp_WdfIoTargetStop+0xba 804716c4 f39713d2 7c9b1968 00000003 54434163 xxx!WdfIoTargetStop+0x1a [...kmdf10\inc\wdfiotarget.h @ 375] 80471700 f38d2272 7c9b1968 7c9c3ff0 00000003 xxx!xxxEvtBulkInCompletion+0xe2 [c:\xxx\readwrit.c @ 326] 80471730 f38f270d 7c97a1c8 7c9b1968 83653b14 Wdf01000!FxUsbPipeContinuousReader::_FxUsbPipeRequestComplete+0x3e 80471758 f38cb1dc ae08bff8 8364e690 00000000 Wdf01000!FxRequestBase::CompleteSubmitted+0x89 80471774 f38cb296 01685e30 ae08bf00 804717cc Wdf01000!FxIoTarget::RequestCompletionRoutine+0x195 80471784 80528ee6 00000000 ae08bf00 83685e30 Wdf01000!FxIoTarget::_RequestCompletionRoutine+0x35 804717cc f742e8ae ae08bf00 83679568 85e39028 nt!IovSpecialIrpCompleteRequest+0x18c 80471820 f742e5a1 46616d64 43707269 00000000 USBPORT!USBPORT_CompleteTransfer+0x306 80471850 f742fdcd 546e6f02 80464224 83651708 USBPORT!USBPORT_DoneTransfer+0x135 8047187c f74324c8 85e39028 80464224 85e391f0 USBPORT!USBPORT_FlushDoneTransferList+0x15f 804718ac f74401a1 85e39028 80464224 85e39028 USBPORT!USBPORT_DpcWorker+0x2d2 804718f4 f7440309 85e39028 00000001 8046d6f0 USBPORT!USBPORT_IsrDpcWorker+0x36f 80471910 804644c7 85e39608 6b755044 00000000 USBPORT!USBPORT_IsrDpc+0x15b 80471924 80464462 0000000e 00000000 00000000 nt!KiRetireDpcList+0x30 ffdff800 85e3962c 00000001 80471928 0006e721 nt!KiIdleLoop+0x26 WARNING: Frame IP not in any known module. Following frames may be wrong. ffdff804 00000000 80471928 0006e721 0006e718 0x85e3962c Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84796
 - 11
 - Some basic Unicode String questions Hello, I have some questions about Unicode Strings. I looked over MSDN and I couldn't find anything. My last choice was to go through the source code, but I would find it better to ask someone here. 1) Does RtlInitUnicodeString allocate a buffer for me to use? So if I do: UNICODE_STRING myUnicodeString; RtlInitUnicodeString( myUnicodeString, L"hello"); does this allocate room for my unicodeString? 2) I have been initializing my UNICODE_STRING manually, but it seems to get a bit tedious. The way I have been initializing them is like: UNICODE_STRING myString; WCHAR buf[255]; RtlZeroMemory( buf, 255); myString.Buffer = buf; myString.MaximumLength = 255; myString.Length = myString.MaximumLength but this seems long, isn't there a better way? 3) Also, what is the difference between doing this: UNICODE_STRING myUnicodeString; RtlInitUnicodeString( myUnicodeString, L"hello"); and... UNICODE_STRING myUnicodeString; CCHAR ntNameFile[12] = "hello"; STRING ntNameString; RtlInitAnsiString( &ntNameString, ntNameFile); RtlAnsiStringToUnicodeString(&myUnicodeString, &ntNameString, TRUE ); Thanks, any help would be appreciated.. Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84794
 - 12
 - USB hub reduces transfer speed? Hi, I am developing a USB driver for a device that reads continous data from the ftdi 245 chip. I am requesting 131072 bytes of data per UsbBuildInterruptOrBulkTransferRequest. This works nice when i am running with the device inserted in on of the laptop computer USB ports, but as soon as i use any 2.0 or 1.1 or 1.0 USB hub, the amount of data returned from every transfer varies from 200 - 50000 bytes, never the amount i requested. Why is it so, that when i use a USB hub, my data transfer slows down alot? Running XP sp2, fully updated. Same thing happens on similar computers, though on my stationary computer with XP sp2, everything works just fine. Any ides or suggestions are very welcome. Regards Wilhelm Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84792
 - 13
 - Memory ordering and flag/state variables If a flag (in a device extension for instance) is set in some driver routine and checked in another, shouldn't the first routine technically perform some sort of memory barrier (KeMemoryBarrier() or an InterlockedExchange()) to make sure the new value of the flag would be seen by all CPUs before the function returns? This isn't the same as a test-and-set. One thread only sets the flag and the other only tests the flag, but when the test succeeds the results of any other operations indicated by the flag should be finished. I found out about memory ordering a little while ago, and I've heard the theories about how out-of-order loads and stores can cause problems, but when I look at a lot of code I don't see as many barriers as I'd expect and I wonder why it works. I often just see plain old C assignment statements, and I wonder if it happens to work either because current CPUs order their memory operations or because there are already so many implicit barriers from other intervening operations. Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84790
 - 14
 - PNP reentrance Can a driver receive a PNP IRP while its in the middle of processing another PNP IRP? Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84788
 - 15
 - How to find PortName I wanted to know how I can figure out the port name (e.g. COMx) that is assigned to a PCMCIA modem. or how I can figure out the device info from port name (e.g. COMx). Would you mind pasting a sample code? In addition, I wanted to know if I were to write the serial dirver for this modem how I can get this infromation from the kernel mode driver. How is the device assigned a COMx number? Does the driver assign the port name or is this done by the OS? Shreyash B. Patel Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84787
 - 16
 - No waiting in IRP_MN_REMOVE_DEVICE? In the IRP_MN_REMOVE_DEVICE examples in the WDM book and the DDK, near the end the IRP is forwarded to the driver below it and immediately after this (without waiting) the upper device object is detached and deleted. What made me curious is why there's no need to wait for the lower driver to complete the IRP. Does the I/O manager never have a need to touch the device object again even as the IRP moves back up the stack during completion, or does this happen to work because IRP_MN_REMOVE_DEVICE will complete immediately and never be pended? Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84782
 - 17
 - Using WPP in kernel mode drivers that are build using VS/VS2005 Hello, I have a kernel mode driver project and use VS2005 to build this driver. I want to generate logs from this driver and want to use WPP for that reason. I am having diffuclty in defining RUN_WPP preprocessor within VS2005 build settings. If anyone has done similar from VS2005, please part your knoweldge with me. Thanks Brijesh Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84778
 - 18
 - WttLogger with DTM Hi, I am new in here and I'm currently configuring my tests to run with DTM. Unfortunatly, I cannot access the windows.beta.wdk.dtm newsgroup due to access problem :( My question is: Where can I find the WTTLogger.h and .lib ? I need them to use the CWTTLogger class within my tests. Anyone started using DTM yet ? Thank you Sylvain Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84777
 - 19
 - Whats with red x's in DDK Build 1830? I needed to update my DDK to 1830 and notice all of the red x's which mean that these marked features are not available. No audio stuff, most samples are not available, win xp headers, etc.. What am I missing here? I dont recall my 3790 doing this at all. Thanks, Mike Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84776
 - 20
 - 1394B S800 Dear sir, When will the Microsoft Software driver support S800 in 1394B PCI card ? Where can I find the S800 driver for 1394B PCI card ? best regards, David Tang Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84763
 - 21
 - 1394 DDI document Dear sir, (1) where can I find the "1394 DDI document" ? (2) How to enable BOOT UP with "1394 diag mode" ? (I do not want to use UI to start the "1394 diag mode") Best regards, Please, help me!! Thanks, David Tang Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84762
 - 22
 - Newbie Question : PnP & PS2 ? AFAIK there is no PnP support for PS2 interface in Windows kernel , this is because of PS2 hw or not ? Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84754
 - 23
 - File System - large file random access Hi, Is it possible to insert data in the middle of a file without affecting other data? I need this to improve system performance. Assume a large database file (100MB). Now I am inserting 10 bytes in the middle of the file(50MB). What the actions will be taken by the file system? I am assume one of the following action will be taken 1) Insert the new 10 bytes at the 50MB offset and move the bottom half of the 50MB data 10 byte below and allocating a new disk block in the end if required. 2) Allocating new disk block in the middle for the 10 byte and just doing the re-linking. Method 2 will be faster because it requires less number of disk accesses for large files. Is there any file system supporting this kind of functionality TIA, Samuel Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84752
 - 24
 - Getting STATUS_TOO_LATE with virtual disk I have a driver that creates a virtual NTFS formatted removable disk drive that is actually just a file on an NTFS formatted hard disk. Each time the system is shutdown, NTFS wants to write data to the virtual disk. Usually, writes succeed but at shutdown calling ZwWriteFile to carry out the write requests results in STATUS_TOO_LATE. Is there a way to resolve this? Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84749
 - 25
 - Are VISTA RC1 and WDK RC1 the same ISO file? Hi, I am new here,and I am an engineer for our products driversâ?? WHQL testing. Our drivers had passed HCT testing and got microsoft logo'd driver. However, we need to transfer to DTM testing tool to retest our drivers. So I attend Microsoft connect program, and I can download WDK RC1 from Microsoft connect . And I also heard of VISTA RC1 has released now, Iâ??m wondering that WDK RC1 and VISTA RC1 are the same ISO file? If not, where can I download VISTA RC1? I think I should test our drivers on VISTA platform asap. Can anyone help me out ? Thanks! Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84747
- Next
 - 1
 - determine if using 802.1x Anyone know if it is possible to determine programatically if a connection that I am using, used 802.1x to authenticate, when connecting to my switch? --joubert Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84746
 - 2
 - WinCE modifications Serial_class to USBser class Anybody can help to modify my existing serial class to USBser class for USBser.sys support for XP Host, Please help me, who are using Windows Mobile 5.0 or Windows CE 6.0 beta this two platform have full support of USBser.sys. for USB modem class on Win CE device. for refrance please check it out http://groups.google.com/group/microsoft.public.windowsce.embedded.vc/browse_thread/thread/cc1542edb42abc63/4135e4aa3d883c8e?lnk=st&q=&rnum=17&hl=en#4135e4aa3d883c8e I able to connect device by modifying in descriptor but I am still not able to open COM port in Hyperterminal to test...:( in Windows Mobile 5 platform builder WINCE500\public\common\oak\drivers\usbfn\class\serial ---->USBFnDrv.cpp in Windows CE 6 beta may be they have give seprate class as WINCE500\public\common\oak\drivers\usbfn\class\USBser I appreciate who would give me the direction. Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84743
 - 3
 - Checking if path is valid. Hello, I am creating a keyboard filter logger as a learning experience. So far, everything is going pretty smooth, and I think i am understanding the basics. I am having one problem though. I allow the user to specify the path of where he or she wants the log file in the registry. With this value, I will be calling NtCreateFile to create the log file and write it etc.. Is there a way to validate the path though? I was thinking about calling NtOpenFile to see if it fails and if it fails, then I could tell if the path that is provided in the registry is broken, but that seems a bit dirty. I was hoping to find a clean way to do it. Also, if I pass in "c:\\log.txt" as opposed to "\\DosDevices\\c:\\log.txt" to NtCreateFile, would it still work properly? or must I pass in the full path. thanks Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84742
 - 4
 - IoRegisterDeviceInterface outside of AddDevice I would like to register a different device interface GUID depending on my PCI Device ID (using the same binary WDM driver). 1. Is it OK to wait until the IRP_MN_START_DEVICE minor function to call IoRegisterDeviceInterface? I don't want to access the device until I know it's started, and I can't select the proper interface to register until I know which device I've got... 2. If it is not OK to register the device interface then, is there a way to get my Hardware ID when I'm in the AddDevice routine? Thanks Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84737
 - 5
 - Extra ObReferenceObject() -- is this why? I think I almost understand what this statement from the WDM book means (in reference to sending an asynchronous IRP to another driver): "The calls to ObReferenceObject and ObDereferenceObject that precede and follow the call to IoCallDriver ... are necessary only when you've used IoGetDeviceObjectPointer to obtain the DeviceObject pointer and when the completion routine (or something it calls) will release the resulting reference to a device or file object." I can only think that this also assumes there's code immediately after IoCallDriver() that accesses the device object (for some reason I can't think of) and may end up running either before or after the completion routine. If there is no code like that then I'd think the reference from IoGetDeviceObjectPointer() and the dereference from the completion routine would pair with each other and be safe. The book's examples are all in this pseudo-code style with missing details, and it doesn't actually show whether code like that would or wouldn't be there. I don't see any similar real examples in the included CD samples. Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84733
 - 6
 - Getting a value from the registry. Hello, I am having problems reading a value from the registry... For some reason, the return value is always empty. Would anyone be willing to glance over my code? reg_key is the key reg_path is the path to the registry which is passed in from Driver Entry - Registry Path val is the out value that will store the result. thanks ------------------------------------------------------ NTSTATUS GetRegistryValue( IN UNICODE_STRING reg_key, IN PUNICODE_STRING reg_path, OUT PUNICODE_STRING val ) { NTSTATUS status; HANDLE hkey; OBJECT_ATTRIBUTES obj_attrib_reg; ULONG size = 0; // Opening the registry to query it InitializeObjectAttributes( &obj_attrib_reg, reg_path, (OBJ_KERNEL_HANDLE | OBJ_CASE_INSENSITIVE), NULL, NULL ); status = STATUS_UNSUCCESSFUL; status = ZwOpenKey(&hkey, KEY_READ, &obj_attrib_reg); // If it is successful, it means that we were able to open the registry value and // that there exists a registry key with the name provided above. if( NT_SUCCESS(status)) { status = STATUS_UNSUCCESSFUL; status = ZwQueryValueKey( hkey, &reg_key, KeyValuePartialInformation, NULL, 0, &size); // If we were able to successfully query for the key, then the key is valid. // This does not mean that the value was empty. if( !(status == STATUS_OBJECT_NAME_NOT_FOUND || size == 0 )) { size = min(size, PAGE_SIZE); PKEY_VALUE_PARTIAL_INFORMATION vpip = (PKEY_VALUE_PARTIAL_INFORMATION)ExAllocatePool(PagedPool, size); if( !vpip ) { ZwClose(hkey); return STATUS_UNSUCCESSFUL; } status = STATUS_UNSUCCESSFUL; status = ZwQueryValueKey( hkey, &reg_key, KeyValuePartialInformation, vpip, size, &size); if(!NT_SUCCESS(status)) { ZwClose(hkey); return STATUS_UNSUCCESSFUL; } // Copying my data into the string so that I could use it when I return PCANSI_STRING my_data = (PCANSI_STRING)vpip->Data; RtlAnsiStringToUnicodeString( val, my_data, TRUE); ExFreePool(vpip); } else { ZwClose(hkey); return STATUS_UNSUCCESSFUL; } ZwClose(hkey); } else { return STATUS_UNSUCCESSFUL; } return STATUS_SUCCESS; } Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84731
 - 7
 - Byte array of each page from the printer driver I have implemented a virtual printer driver using sample Microsoft unidrv approch. I want to split the pages so that I can get the byte array of each page individually. After all pages byte array I recieves, I need to call a .Net application which process these pages one by one & send data to our customize application. Thanks, Sahaj Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84730
 - 8
 - DMA abstraction for x64 drivers / performance issues This is a multi-part message in MIME format. ------=_NextPart_000_0017_01C6D0E3.D7A869D0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Considering page 9 of the following document :=20 http://www.microsoft.com/whdc/driver/kernel/dma.mspx and dealing with 64-bit adapters used in x64 machines that will have at = least 8GB RAM, featuring the AMD64 IOMMU implementation (as opposed to = bouncing buffers above 4GB), I wonder if opting for DMA abstraction = could tax performance by more than 30% in some cases. Below is Microsoft's rationale :=20 The Windows PHYSICAL_ADDRESS data structure has always been 64 bits. If = you expect an address that can be safely truncated to 32-bits, it is = good programming practice to ASSERT that the high half of the address is = zero. =B7 The bus addressing capability. Driver writers that follow = the Windows DMA model never have to be concerned about the addressing = capability of the bus to which their device is attached. For example, = the driver for a PCI bus device that is capable of 64-bit addressing = does not need to be aware of whether the device is attached to a 64-bit = capable PCI bus. Just as with the amount of physical memory present on = the machine, the addressing capability of the bus is made transparent by = the standard Windows DMA implementation. Bus addressing transparency is maintained for PCI buses with 64 address = lines, as well as for PCI buses that support the dual address cycle = (DAC) mode of 64-bit addressing. =20 The above features apply only to drivers that follow the model based on = the Windows DMA abstraction. Drivers that take shortcuts by not = following the model, often in an unnecessary attempt at performance = optimization, will receive unpredictable levels of support for the = features described above, depending on precisely which parts of the DMA = model they use. Thanks,=20 Christopher ------=_NextPart_000_0017_01C6D0E3.D7A869D0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> <HTML><HEAD> <META http-equiv=3DContent-Type content=3D"text/html; = charset=3Diso-8859-1"> <META content=3D"MSHTML 6.00.2900.2963" name=3DGENERATOR> <STYLE></STYLE> </HEAD> <BODY> <DIV>Considering page 9 of the following = document :=20 </DIV> <DIV> <A=20 href=3D"http://www.microsoft.com/whdc/driver/kernel/dma.mspx">http://www.microsoft.com/whdc/driver/kernel/dma.mspx</A><= /P> and dealing with 64-bit adapters used in x64 machines that = will have=20 at least 8GB RAM, featuring the AMD64 IOMMU implementation (as = opposed=20 to bouncing buffers above 4GB), I wonder if opting for DMA=20 abstraction could tax performance by more than 30% in some=20 cases. Below is Microsoft's rationale : The Windows <SPAN=20 style=3D"FONT-WEIGHT: normal">PHYSICAL_ADDRESS data = structure has=20 always been 64 bits. If you expect an address that can be safely = truncated to=20 32-bits, it is good programming practice to ASSERT that the high half of = the=20 address is zero. <FONT=20 size=3D2>=B7<SPAN=20 style=3D"FONT: 7pt 'Times New = Roman'">        =20 The bus addressing capability. Driver writers that follow = the=20 Windows DMA model never have to be concerned about the addressing = capability of=20 the bus to which their device is attached. For example, the driver for a = PCI bus=20 device that is capable of 64-bit addressing does not need to be aware of = whether=20 the device is attached to a 64-bit capable PCI bus. Just as with the = amount of=20 physical memory present on the machine, the addressing capability of the = bus is=20 made transparent by the standard Windows DMA = implementation. Bus addressing transparency is maintained for PCI buses with 64 = address=20 lines, as well as for PCI buses that support the dual address cycle = (DAC) mode=20 of 64-bit addressing. <FONT=20 size=3D2> <?xml:namespace prefix =3D o ns =3D=20 "urn:schemas-microsoft-com:office:office" = /><o:p></o:p> The above features apply only to drivers that follow the model = based on=20 the Windows DMA abstraction. Drivers that take shortcuts by not = following the=20 model, often in an unnecessary attempt at performance optimization, will = receive=20 unpredictable levels of support for the features described above, = depending on=20 precisely which parts of the DMA model they use.   Thanks,=20 Christopher  </DIV></BODY></HTML> ------=_NextPart_000_0017_01C6D0E3.D7A869D0-- Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84721
 - 9
 - Computer Crash - Device Driver I own a Gateway 500X puter purchased in June 2003. Recently I have been experiencing complete shutdowns - two days ago 5 times. Here is the message error I received: "You are receiving this message because a device driver installed on your computer caused a stop error message. This error message required a restart of your computer. After the restart your computer should continue to operate normally, however the error may re-occur until the problem is corrected. A stop error occurs when your computer encounters an error from which it cannot recover. This is usually caused by a device driver which encounters an unhandled exception or performs an illegal operation. When the operating system detects this situation it is stopped to prevent future problems, such as data loss or further system instability" When I purchased the 500X a red Drivers CD came with the unit. The question I have is should I reinstall the drivers using this CD and will this correct the problem? Or is there another solution out there? Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84720
 - 10
 - Windows 2000 not reloading the driver Hello, Im having a wierd issue in Windows 2000, does not appear in XP. I insert my USB 1.1 device, install the drivers, and drivernetry and adddevice and everything get called and the drivers gets loaded nicely. If i remove the device unload is called. Now begins the strange part. When i reinsert the device, nothing at all happens. Sometimes windows tells me it found a usb device, not recognised and no function driver loaded. Sometimes it just shows me usb device in device manager and nothing is loaded. Here is the even stranger part, if i do the same action, but via a usb hub (unpowered), it works like a charm. Ive been trying to fix this issue several days now, buit im not any closed. What could cause such a strange behavior? Please give me some suggestions! Regards Wilhelm Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84705
 - 11
 - DIFx and virtual devices Hi! I want to install a virtual system device (root\MyBusDevice) for x64 systems. Is that possible using the DIFx tools? Cause I am getting these DIFx logs: DIFXAPP: INFO: Looking for Model Section ... DIFXAPP: INFO: No drivers installed. No devices found that match driver(s) When I use the DevCon sample tool from the DDK it works for x64 systems. I just read that the Vista SDK includes a new version of the DIFx tools -> Version 2.10. I will give it a try. Thanks Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84701
 - 12
 - Help Needed Good Morning to All, I am working on File System Development for an Embedded OS to NandFlash. I need to know about the FAT file system concepts, its working principles, interaction between File System and File system Driver. If you know any books or online resources pls send me. I also have a query : How does the OS able to find the size of the device when formatting? Like floppy and Harddisk. Are these information will be stored in the device itself? Thank you Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84689
 - 13
 - Where do I get Signability Test? Hi, I have tried to download Signability Test tool from the following site: Download WHQL Tests, Procedures and Forms http://www.microsoft.com/whdc/whql/WHQLdwn.mspx Signability Test https://winqual.microsoft.com/com/mercsetup.exe However, the link for the tool seems dead and I could not download it. Does anyone know how to get this tool? Especially, I would like to obtain the newest version (v2.19). (I know it is included in WDK, but I need separated package as well.) Thanks Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84687
 - 14
 - Driver prevent S1 standby but aloow S3 standby ? Hi, Our device driver is a USB port driver sitting below a mouse class driver. I found if I set the BIOS option "Suspend to RAM enabled (S3)".. WinXP is able to go to standby with our deivce attached. While if BIOS option "Suspend to RAM disabled(S1)".. system would claim that our driver prevent the system going to standby.. The following is a bit of debug info... I found our driver return success for Query Power,, but system just set the device to D0..when S1 .. ,, (while it would set device to D3 when S3 Query..) Any hint for that ? Thanks. -chueh8 TSC(IRQL 00): TSCPowerDispatch++. IRP 8273AE48 TSC(IRQL 00): IRP 8273AE48 IRP_MJ_POWER TSC(IRQL 00): IRP_MN_QUERY_POWER TSC(IRQL 00): PowerSystemSleeping1 TSC(IRQL 00): ............Query Power 8273ae48.... TSC(IRQL 00): query system power TSC(IRQL 00): TSCSystemPowerDispatch++. IRP 8273AE48 TSC(IRQL 00): TSCSystemPowerCompletionRoutine++. IRP 8273AE48 STATUS 0 TSC(IRQL 00): KKK1 TSC(IRQL 00): TSCGetDevicePowerState++. PowerSystemSleeping1 TSC(IRQL 00): System State =2 TSC(IRQL 00): DeviceExtension->DeviceCaps.SystemWake =4 TSC(IRQL 00): DeviceExtension->DeviceCaps.DeviceState[SystemState] =3 TSC(IRQL 00): DeviceExtension->DeviceCaps.WakeFromD3 =1 TSC(IRQL 00): DeviceExtension->DeviceCaps.WakeFromD2 =1 TSC(IRQL 00): DeviceExtension->DeviceCaps.WakeFromD1 =1 TSC(IRQL 00): DeviceExtension->DeviceCaps.WakeFromD0 =1 TSC(IRQL 00): 333 TSC(IRQL 00): TSCGetDevicePowerState--. PowerDeviceD3 STATUS 0 TSC(IRQL 00): Get Device Power state status=0 TSC(IRQL 00): KKK2 TSC(IRQL 00): TSCDevicePowerCompleteCallback++. IRP 8273AE48 TSC(IRQL 00): TSCDevicePowerCompleteCallback-- TSC(IRQL 00): KKK4 TSC(IRQL 00): TSCSystemPowerCompletionRoutine--. Irp 8273AE48 STATUS 103 TSC(IRQL 00): TSCSystemPowerDispatch--. IRP 8273AE48 TSC(IRQL 00): TSCPowerDispatch--. IRP 8273AE48 STATUS 103 TSC(IRQL 00): TSCPowerDispatch++. IRP 8273A008 TSC(IRQL 00): IRP 8273A008 IRP_MJ_POWER TSC(IRQL 00): IRP_MN_SET_POWER TSC(IRQL 00): PowerSystemWorking TSC(IRQL 00): ............PDO set power 1 .... TSC(IRQL 00): set sys power TSC(IRQL 00): TSCSystemPowerDispatch++. IRP 8273A008 TSC(IRQL 00): TSCSystemPowerCompletionRoutine++. IRP 8273A008 STATUS 0 TSC(IRQL 00): KKK1 TSC(IRQL 00): TSCGetDevicePowerState++. PowerSystemWorking TSC(IRQL 00): System State =1 TSC(IRQL 00): DeviceExtension->DeviceCaps.SystemWake =4 TSC(IRQL 00): DeviceExtension->DeviceCaps.DeviceState[SystemState] =1 TSC(IRQL 00): DeviceExtension->DeviceCaps.WakeFromD3 =1 TSC(IRQL 00): DeviceExtension->DeviceCaps.WakeFromD2 =1 TSC(IRQL 00): DeviceExtension->DeviceCaps.WakeFromD1 =1 TSC(IRQL 00): DeviceExtension->DeviceCaps.WakeFromD0 =1 TSC(IRQL 00): 111 TSC(IRQL 00): TSCGetDevicePowerState--. PowerDeviceD0 STATUS 0 TSC(IRQL 00): Get Device Power state status=0 TSC(IRQL 00): KKK2 TSC(IRQL 00): TSCPowerDispatch++. IRP 82440430 TSC(IRQL 00): IRP 82440430 IRP_MJ_POWER TSC(IRQL 00): IRP_MN_SET_POWER TSC(IRQL 00): PowerDeviceD0 TSC(IRQL 00): ............PDO set power 1 .... TSC(IRQL 00): set device power TSC(IRQL 00): device same power TSC(IRQL 00): device same power 2 TSC(IRQL 00): TSCPowerD0PrepCallback++. IRP 82440430 TSC(IRQL 00): TSCDevicePowerCompleteCallback++. IRP 8273A008 TSC(IRQL 00): TSCDevicePowerCompleteCallback-- TSC(IRQL 00): TSCPowerD0PrepCallback-- TSC(IRQL 00): TSCPowerDispatch--. IRP 82440430 STATUS 103 TSC(IRQL 00): KKK4 TSC(IRQL 00): TSCSystemPowerCompletionRoutine--. Irp 8273A008 STATUS 103 TSC(IRQL 00): TSCSystemPowerDispatch--. IRP 8273A008 TSC(IRQL 00): TSCPowerDispatch--. IRP 8273A008 STATUS 103 Abort: \Driver\Mouclass \Device\PointerClass1 irp (8273ae48) QueryPower-Sleeping1 status c00002d3 Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84683
 - 15
 - Need guidance on Generic USB IO (by using PipeHandles??) I am a Win32 application programmer and a newbie to USB driver development. I am stuck with the changed requirements for my application, please guide me on this. About the Application: 1. The application is intended to give a solution to control most USB devices, mainly printers(Inluding DOT4 printers). 2. The communication with USB devices is implemented as a Win32 User DLL which the Application(Java) will load dynamically and communicate with the USB device. 3. This DLL uses the Connection Address(Port Address) and uses it to fetch the corresponding Symbolic Link(Device Path). The symbolic link is sent to the CreateFile() to obtain the Device Handle, which is finally used for further communication with the USB device using ReadFile()/WriteFile(). 4. The application currently uses only the drivers that are shipped with Windows OS. And no custom drivers are used as of now for USB communication. Current requirement: 1. Now to make the application more generic so that it can control other USB devices apart from USB printers, the idea is to use the "Pipe handle" for data exchange instead of "Device Handle" from the Win32 DLL. Queries: 1. Can we get the handle to a pipe by some means(append the pipe name to the Symbolic link??) for communication? 2. Can we use pipeHandle straight from a Win32 Application that uses existing USB drivers shipped with Windows? 3. Does this requirement(Using PipeName for USB data exchange) need any custom driver or can this be achieved using the USB drivers that are shipped with Windows? Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84647
 - 16
 - newbie@KMDF: can not build a signed catalog file for Vista 64. Hello everybody! *** Problem: WDF/KMDF: I can not build a signed catalog file (signtool.exe verify gives errors). I use already the updated signtool.exe (Version 6.0.5440.0). - signtool.exe verify gives an undocumented error - Vista says about my inf-file: "This driver is not digitally signed!" So I think to whole signing process failed *** Background: Because I worked the last years with Compuware Driver:Works, I'm quite new to the driver development process with Microsoft tools. I now have to port a driver to WDF/KMDF and try to get familiar with the driver signing process for Vista. For test of my own build environment, I want to get the WDF-ECHO-example running on Vista on a AMD64x2 machine. For build, I use WindowsXP. *** I list here all steps of my building and signing process: 1) This is how I made a test certificate: CMD>c:\WinDDK\5384\bin\selfsign\Makecert.exe -r -pe -ss PrivateCertStore -n "CN=PEAKTESTCERT" D:\PEAK\PEAKCAN3\testcert.cer ... seems OK. 2) This is how I build ECHO (copied into my own build directory): CMD>call C:\WinDDK\5384\bin\setenv.bat c:\WinDDK\5384\ CHK x86 WLH CMD>cd /d D:\PEAK\PEAKCAN3\CURRENT\ECHO CMD>build -ceZP ... seems OK. Now I have echo.sys and echo.inf. See echo.inf at the end of thsi email! 3) Now I copy the coinstaller DLL in the directory where echo.sys and echo.inf were generated: CMD>copy c:\WinDDK\5384\redist\wdf\x86\WdfCoInstaller01005.dll D:\PEAK\PEAKCAN3\CURRENT\ECHO\sys\OBJCHK_WLH_x86\i386 4) I stamp ECHO.INF: CMD>c:\WinDDK\5384\bin\x86\stampinf -f ECHO.inf -d 09/04/2006 -v 6.0.9999.0 ... seems OK 5) Now I create an catalog file, I call it "KmdfSamples.inf" I use "makecat.exe" and a *.cdf-File, see below. (This is because SIGNABILITY.EXE does not work at all ...) ----------------- BEGIN pack_sign.cdf -------------------------------- [CatalogHeader] Name=KmdfSamples ResultDir=D:\PEAK\PEAKCAN3\CURRENT\ECHO\sys\OBJCHK_WLH_x86\i386 [CatalogFiles] <HASH>wdfcoinstaller=D:\PEAK\PEAKCAN3\CURRENT\ECHO\sys\OBJCHK_WLH_x86\i386\WdfCoInstaller01005.dll <HASH>inffile=D:\PEAK\PEAKCAN3\CURRENT\ECHO\sys\OBJCHK_WLH_x86\i386\ECHO.inf <HASH>sysfile=D:\PEAK\PEAKCAN3\CURRENT\ECHO\sys\OBJCHK_WLH_x86\i386\ECHO.sys --------------------- END pack_sign.cdf -------------------------------- CMD>c:\WinDDK\5384\bin\selfsign\makecat.exe -v D:\PEAK\PEAKCAN3\Scripts\execute\pack_sign.cdf opened: D:\PEAK\PEAKCAN3\Scripts\execute\pack_sign.cdf processing: <HASH>wdfcoinstaller processing: <HASH>inffile processing: <HASH>sysfile Succeeded 6) I try to sign my catalog file "KmdfSamples.inf" CMD>c:\WinDDK\5384\bin\selfsign\SignTool.exe sign /v /s PrivateCertStore /n PEAKTESTCERT D:\PEAK\PEAKCAN3\CURRENT\ECHO\sys\OBJCHK_WLH_x86\i386\KmdfSamples.cat The following certificate was selected: Issued to: PEAKTESTCERT Issued by: PEAKTESTCERT Expires: 01.01.2040 01:59:59 SHA1 hash: 91F8CF9553ED09B76D039A17646416533D9534D0 Done Adding Additional Store Attempting to sign: D:\PEAK\PEAKCAN3\CURRENT\ECHO\sys\OBJCHK_WLH_x86\i386\KmdfSamples.CAT Successfully signed: D:\PEAK\PEAKCAN3\CURRENT\ECHO\sys\OBJCHK_WLH_x86\i386\KmdfSamples.CAT Number of files successfully Signed: 1 Number of warnings: 0 Number of errors: 0 ... it seems OK 7) I timestamp the catalog file CMD>c:\WinDDK\5384\bin\selfsign\SignTool.exe timestamp /v /t http://timestamp.verisign.com/scripts/timestamp.dll D:\PEAK\PEAKCAN3\CURRENT\ECHO\sys\OBJCHK_WLH_x86\i386\KmdfSamples.cat Timestamping: D:\PEAK\PEAKCAN3\CURRENT\ECHO\sys\OBJCHK_WLH_x86\i386\KmdfSamples.CAT Successfully timestamped: D:\PEAK\PEAKCAN3\CURRENT\ECHO\sys\OBJCHK_WLH_x86\i386\KmdfSamples.CAT Number of files successfully timestamped: 1 Number of errors: 0 ... it seems OK 8) I verify the catalog file: CMD>c:\WinDDK\5384\bin\selfsign\SignTool.exe verify /v /pa D:\PEAK\PEAKCAN3\CURRENT\ECHO\sys\OBJCHK_WLH_x86\i386\KmdfSamples.cat Verifying: D:\PEAK\PEAKCAN3\CURRENT\ECHO\sys\OBJCHK_WLH_x86\i386\KmdfSamples.CAT Signing Certificate Chain: Issued to: PEAKTESTCERT Issued by: PEAKTESTCERT Expires: 01.01.2040 01:59:59 SHA1 hash: 91F8CF9553ED09B76D039A17646416533D9534D0 The signature is timestamped: 04.09.2006 10:38:20 Timestamp Verified by: Issued to: Thawte Timestamping CA Issued by: Thawte Timestamping CA Expires: 01.01.2021 01:59:59 SHA1 hash: BE36A4562FB2EE05DBB3D32323ADF445084ED656 Issued to: VeriSign Time Stamping Services CA Issued by: Thawte Timestamping CA Expires: 04.12.2013 01:59:59 SHA1 hash: F46AC0C6EFBB8C6A14F55F09E2D37DF4C0DE012D Issued to: VeriSign Time Stamping Services Signer Issued by: VeriSign Time Stamping Services CA Expires: 04.12.2008 01:59:59 SHA1 hash: 817E78267300CB0FE5D631357851DB366123A690 Number of files successfully Verified: 0 Number of warnings: 0 Number of errors: 1 *** What is this error??? 9) I try to install on Vista: After I select the echo.inf, there's the message "This driver is not digitally signed!" So I think to whole signing process failed ----------------- BEGIN echo.inf ----------------------- ;/*++ ; ;Copyright (c) 1990-2000 Microsoft Corporation ; ;Module Name: ; ECHO.INF ; ;Abstract: ; INF file for installing the Driver Frameworks ECHO Driver ; ;Author: ; Ravi Gollapudi (RaviG) Oct 2002 ; ;Installation Notes: ; Using Devcon: Type "devcon install ECHO.inf root\ECHO" to install ; ;--*/ [Version] Signature="$WINDOWS NT$" Class=Sample ClassGuid={78A1C341-4539-11d3-B88D-00C04FAD5171} Provider=%MSFT% DriverVer=09/04/2006,6.0.5384.4 CatalogFile=KmdfSamples.cat [DestinationDirs] DefaultDestDir = 12 ; ================= Class section ===================== [ClassInstall32] Addreg=SampleClassReg [SampleClassReg] HKR,,,0,%ClassName% HKR,,Icon,,-5 [SourceDisksNames] 1 = %DiskId1%,,,"" [SourceDisksFiles] ECHO.sys = 1,, ;***************************************** ; ECHO Install Section ;***************************************** [Manufacturer] %StdMfg%=Standard,NTx86 ; Following section is meant for Windows 2000 as it ; cannot parse decorated model sections [Standard] ; ; Hw Id is root\ECHO ; %ECHO.DeviceDesc%=ECHO_Device, root\ECHO ; Decorated model section take precedence over undecorated ; ones on XP and later. [Standard.NTx86] %ECHO.DeviceDesc%=ECHO_Device, root\ECHO [ECHO_Device.NT] CopyFiles=Drivers_Dir [Drivers_Dir] ECHO.sys ;-------------- Service installation [ECHO_Device.NT.Services] AddService = ECHO,%SPSVCINST_ASSOCSERVICE%, ECHO_Service_Inst ; -------------- ECHO driver install sections [ECHO_Service_Inst] DisplayName = %ECHO.SVCDESC% ServiceType = 1 ; SERVICE_KERNEL_DRIVER StartType = 3 ; SERVICE_DEMAND_START ErrorControl = 1 ; SERVICE_ERROR_NORMAL ServiceBinary = %12%\ECHO.sys LoadOrderGroup = Extended Base ; ;--- ECHO_Device Coinstaller installation ------ ; [DestinationDirs] ECHO_Device_CoInstaller_CopyFiles = 11 [ECHO_Device.NT.CoInstallers] AddReg=ECHO_Device_CoInstaller_AddReg CopyFiles=ECHO_Device_CoInstaller_CopyFiles [ECHO_Device_CoInstaller_AddReg] HKR,,CoInstallers32,0x00010000, "wdfcoinstaller01005.dll,WdfCoInstaller" [ECHO_Device_CoInstaller_CopyFiles] wdfcoinstaller01005.dll [SourceDisksFiles] wdfcoinstaller01005.dll=1 ; make sure the number matches with SourceDisksNames [ECHO_Device.NT.Wdf] KmdfService = ECHO, ECHO_wdfsect [ECHO_wdfsect] KmdfLibraryVersion = 1.5 [Strings] SPSVCINST_ASSOCSERVICE= 0x00000002 MSFT = "Microsoft" StdMfg = "(Standard system devices)" DiskId1 = "WDF Sample ECHO Installation Disk #1" ECHO.DeviceDesc = "Sample WDF ECHO Driver" ECHO.SVCDESC = "Sample WDF ECHO Service" ClassName = "Sample Device" Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84642
 - 17
 - newbie question Hello all, There 2 versions of symbol package: the retail symbols and the checked symbols. What are the differences between both? What are they used for? Thanks for your answers, Jerome Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84641
 - 18
 - unirc.h Hi. I'm trying to build a minidriver using some .gpd, and .rc files. But when I typed "build -cZg", it spat out error saying that I have no unirc.h file, which was to be included in pclxl.rc file. I guess the source files I have were originally written for an older version of ddk, than mine. So I commented the unirc.h-line out and have a good looking .dll output. But I'm not sure if I did properly. My ddk is of ver. 3790.1830. When is unirc.h removed from ddk, and what should be included insteaded of unirc.h, if any? Thankx. Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84640
 - 19
 - USBD_STATUS_CANCELED inconsistent definitios Hello I cannot reply to the other topic about this issue, so here's a new topic... I noticed that the 3790 DDK defines different values for USBD_STATUS_CANCELED in directory <ddk>\inc\ddk\wxp\ like this: usbdi.h: #define USBD_STATUS_CANCELED ((USBD_STATUS)0x00010000L) usb.h: #define USBD_STATUS_CANCELED ((USBD_STATUS)0xC0010000L) My WinXP SP 2 system returns USBD status 0xC0010000 in the URB when my driver cancels the IRP. I don't know about Win2K or WinVista. My driver sources seem to include the 0x00010000 usbdi.h definition. How to deal with this? I also noticed that the usbstor.sys driver masks out bits 31..28 when it checks for USBD status codes like this (I didn't check other system usb drivers): and eax, 0FFFFFFFh cmp eax, 4 <- Apparently checks for USBD_STATUS_STALL_PID This is happening in the real world. So, do we have to mask out the upper 4 bits of every USBD status code before checking it? Is there an incompatibility issue between Win2K, WinXP and WinVista regarding USBD status codes? Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84639
 - 20
 - Memroy mapped files problem in minifiler level Hi, I am writing a filter driver that will encrypt/decrypt the file. The problem i am facing is when some application uses memory mapped files like MS Word or Wordpad to read the file. I cannot intercept the read call for this kind of files, no IRP_MJ_READ for this kinda files. Is there something special i have to do for this kind of memory mapped files. Thanks Lokanadham R Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84634
 - 21
 - Realted with callback interface in Windows drivers Hi All, I am currently working on a layered architecture Windows driver devlopment.My requirement is lower driver must raise a notification to upper layer driver.To do this i think IRP is one of the method... But i wanted to know is there any other way to achieve this , were in upper driver need not register through IRP but still then lower edriver can notify the uppper driver abt the event. Thanks With regards, Jagu Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84631
 - 22
 - InterlockedExchangePointer() while lock held? Is it necessary to do interlocked exchanges while any type of lock is held? The only reason I can think of is to ensure memory ordering, but I got the impression that acquiring and releasing a lock already takes care of that. There's an InterlockedExchangePointer() in a sample in the WDM driver book that accesses a field that looks like it's only ever accessed while a lock is held, and I wondered if the exchange really had to be interlocked in that case or if it was just a lazy way to avoid declaring a temporary variable and typing 3 lines of code. I'm only curious to understand the principle. Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84618
 - 23
 - Simple printer driver Hi, I'm a C/C++ developer looking to create a simple printer driver that outputs data to a text file. I'm working with a proprietary database engine that only allows metadata to be exported from its print function. I would like to output the data to a text file, then use WinDiff on it. Are there any examples that show how to create a print driver like this? Thanks in advance. Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84610
 - 24
 - Problem loading ndis6 light weight filter sample in WDK. Has anyone out there successfully loaded the ndis6 sample filter driver on a recent Vista build? I tried to load the sample on top of my ndis6 miniport driver on Vista 5520 by following the instructions in the code tour but the installer hung and I have to reboot. Before I spend time in debugging the problem, I want to know if there is any gotcha I should have paid attention to. Being relatively unfamiliar with Ndis IM driver, debugging this issue will be difficult since there is absolutely no os symbol for that build. Tom, Max, Stephan or other folks, have you tried that? Thanks in advance. -- Calvin Guan (DDK MVP) Sr. Staff Engineer NetXtreme NTX Miniport Broadcom Corporation Connecting Everything(r) Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84594
 - 25
 - NDIS miniport reset questions Hi everybody, I have some questions with respect to NDIS miniport reset. - In addition to Check4Hang what are the other options to notify NDIS that we are under a reset process? - What would be the impact of the reset on NDIS with respect to the following: - Will it stop transferring packets/OIDS to the miniport? - Will it notify the protocols drivers? - Is there a limit to the time we are allowed to be in reset process? - How would the protocol drivers (especially TCP/IP) behave, in that case? will they stop sending packets/OIDS to NDIS until they are told that the reset is done? Thanks Miki Tag: 4 Full speed isochronous->1 High speed isochronous = 100% CPU Tag: 84578

We are using 4 USB audio device (PCM2902, manufacturer TI) connected to the
Southbridge ( 82801 GB/GR, Manufacturer Intel, a.k.a. ICH7) of a motherboard
through a Multi-TT high speed USB Hub (2507, manufacturer SMSC).

It means that there are four periodical isochronous full-speed streams that
are converted to one High-speed USB stream by the Hub, before being sent to
EHCI controller of the Southbridge.

Software configuration is Windows XP (Professional) SP2 with auto-update
activated. (Kernel Version 5.1.2600.2180)

In this configuration, there is a high increase of CPU load (peak to 100%)
when audio channels are all activated in both direction (Wave In, Wave Out).

It induced lost of many audio frame and lost of audio stream.

Another interesting test has been done connecting directly the four USB
audio devices directly to Southbridge UHCI controller (i.e. using 4
full-speed USB stream).

In this case, it operates correctly (less than 20% CPU load without lost of
audio frame), what validated our software.

Is it a way to use the USB Hub in High speed mode without having the CPU
reaching 100%?

Thanks

Blaise

Top