OSR's ntdev List: Handle Invalid error when trying to run a KMDF driver

Message 1 of 46

26 Feb 08 18:32

ntdev member 38016

xxxxxx@yahoo.com.au

Join Date:

Posts To This List: 23

Handle Invalid error when trying to run a KMDF driver

Hi, I'm trying to develop a virtual kernel-mode device driver for vista that looks like an audio device to the user but under the covers uses winsock kernel to send and receive audio over a network. I've been working on the winsock functions and that builds and runs just fine using the WDK. However, as soon as I add KMDF_VERSION_MAJOR=1 to my sources file, trying to start the driver fails with a "[SC] StartService FAILED 6: The handle is invalid." After fighting with my driver for a while I tried several of the KMDF sample drivers included in the WDK (ndisprot and echo to be precise). I followed the build and installation instructions precisely but both fail to start with the same error message. I have looked in the windows event log, setupact.log, setupapi.app.log and setupapi.dev.log (all with maximum verbosity) (I can post these logs but I don't have attachment permissions). The one lead I have found is a message from wdf01000 in the event log, that can't be resolved: "The description for Event ID 1 from source Wdf01000 cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer." So this is probably the problem, but I cannot find any information about how to reinstall the wdf. I thought that this was the job of the wdf co-installer, but it doesn't update the file (wdf01000.sys). I tried deleting it, after which windows repair put it back again, but that didn't fix my problem. Any thoughts would be much appreciated. Regards, Joe.

Message 2 of 46

26 Feb 08 18:41

Doron Holan

xxxxxx@microsoft.com

Join Date: 08 Sep 2005

Posts To This List: 8270

RE: Handle Invalid error when trying to run a KMDF driver

Which WDK are you using to build your driver? 6000 or 6001? What OS are you trying to install on? It looks like Vista from your description. Is this a PNP driver? If so you cannot use sc.exe to start/stop it. You must device manager or devcon to install the driver and then enable/disable it to force load/unload. If this is a non pnp driver (which is what I am guessing this is based on the samples you mention), there is no INF to run which means the wdf coinstaller does not automatically get into the mix. As such you need to manually invoke the class installer to install the appropriate kmdf runtime. The nonpnp sample (\WinDDK\6000\src\kmdf\nonpnp) shows how to do this in an app d -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of xxxxx@yahoo.com.au Sent: Tuesday, February 26, 2008 3:31 PM To: Windows System Software Devs Interest List Subject: [ntdev] Handle Invalid error when trying to run a KMDF driver Hi, I'm trying to develop a virtual kernel-mode device driver for vista that looks like an audio device to the user but under the covers uses winsock kernel to send and receive audio over a network. I've been working on the winsock functions and that builds and runs just fine using the WDK. However, as soon as I add KMDF_VERSION_MAJOR=1 to my sources file, trying to start the driver fails with a "[SC] StartService FAILED 6: The handle is invalid." After fighting with my driver for a while I tried several of the KMDF sample drivers included in the WDK (ndisprot and echo to be precise). I followed the build and installation instructions precisely but both fail to start with the same error message. I have looked in the windows event log, setupact.log, setupapi.app.log and setupapi.dev.log (all with maximum verbosity) (I can post these logs but I don't have attachment permissions). The one lead I have found is a message from wdf01000 in the event log, that can't be resolved: "The description for Event ID 1 from source Wdf01000 cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer." So this is probably the problem, but I cannot find any information about how to reinstall the wdf. I thought that this was the job of the wdf co-installer, but it doesn't update the file (wdf01000.sys). I tried deleting it, after which windows repair put it back again, but that didn't fix my problem. Any thoughts would be much appreciated. Regards, Joe. --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Message 3 of 46

26 Feb 08 19:38

ntdev member 38016

xxxxxx@yahoo.com.au

Join Date:

Posts To This List: 23

Re: Handle Invalid error when trying to run a KMDF driver

Thanks for your fast reply. In answer to your questions, I tried 6001.17121; when that didn't work, I downloaded and tried 6001.18000. My test platform is vista. It is not a pnp driver; it is a virtual device that looks like an audio device to the user but actually uses winsock kernel to send the audio elsewhere. In trying out echo/ndisprot I followed the instructions in the documentation, including using device manager to start/stop (only they wouldn't start). I have now tried out nonpnp as well, and get what seem to be equivalent symptoms. It asks me to reboot (I doubt this is intended), but after I do and rerun nonpnpapp.exe it just asks me to reboot again. It seems to be tripping over upgrading wdf01000.sys. I have pasted the setupact.log below. Regards, Joe. WdfCoInstaller: [02/27/2008 11:09.54.435] ReadComponents: WdfSection for Driver Service nonpnp using KMDF lib version Major 0x1, minor 0x7 WdfCoInstaller: [02/27/2008 11:09.54.467] DIF_INSTALLDEVICE: Coinstaller version: 1.7.6001 WdfCoInstaller: [02/27/2008 11:09.54.467] GetInMemoryVersionUlong: No information about in-memory KMDF version WdfCoInstaller: [02/27/2008 11:09.54.467] DIF_INSTALLDEVICE: KMDF in-memory version: 0.0.0 WdfCoInstaller: [02/27/2008 11:09.54.560] DIF_INSTALLDEVICE: KMDF on-disk version: 1.5.6000 WdfCoInstaller: [02/27/2008 11:09.54.560] Service Wdf01000 is running WdfCoInstaller: [02/27/2008 11:09.54.560] DIF_INSTALLDEVICE: Reboot is required, so that the newer KMDF version will be loaded to memory WdfCoInstaller: [02/27/2008 11:09.55.122] VerifyMSRoot: exit: error(0) The operation completed successfully. WdfCoInstaller: [02/27/2008 11:09.55.512] Invoking "C:\Windows\system32\wusa.exe "C:\Windows\Temp\WdfTemp\Microsoft Kernel-Mode Driver Framework Install-v1.7-Vista.msu" /quiet /norestart". WdfCoInstaller: [02/27/2008 11:10.17.851] Update process returned error code 0 :error(0) The operation completed successfully. WdfCoInstaller: [02/27/2008 11:10.17.898] InstallComponents: KMDF installed successfully WdfCoInstaller: [02/27/2008 11:10.17.898] InstallComponents: Reboot needed by windows update WdfCoInstaller: [02/27/2008 11:16.19.659] ReadComponents: WdfSection for Driver Service nonpnp using KMDF lib version Major 0x1, minor 0x7 WdfCoInstaller: [02/27/2008 11:16.19.690] DIF_INSTALLDEVICE: Coinstaller version: 1.7.6001 WdfCoInstaller: [02/27/2008 11:16.19.705] GetInMemoryVersionUlong: No information about in-memory KMDF version WdfCoInstaller: [02/27/2008 11:16.19.705] DIF_INSTALLDEVICE: KMDF in-memory version: 0.0.0 WdfCoInstaller: [02/27/2008 11:16.19.705] DIF_INSTALLDEVICE: KMDF on-disk version: 1.5.6000 WdfCoInstaller: [02/27/2008 11:16.19.721] Service Wdf01000 is running WdfCoInstaller: [02/27/2008 11:16.19.721] DIF_INSTALLDEVICE: Reboot is required, so that the newer KMDF version will be loaded to memory WdfCoInstaller: [02/27/2008 11:16.20.095] VerifyMSRoot: exit: error(0) The operation completed successfully. WdfCoInstaller: [02/27/2008 11:16.20.361] Invoking "C:\Windows\system32\wusa.exe "C:\Windows\Temp\WdfTemp\Microsoft Kernel-Mode Driver Framework Install-v1.7-Vista.msu" /quiet /norestart". WdfCoInstaller: [02/27/2008 11:16.41.202] Update process returned error code 0 :error(0) The operation completed successfully. WdfCoInstaller: [02/27/2008 11:16.41.233] InstallComponents: KMDF installed successfully WdfCoInstaller: [02/27/2008 11:16.41.249] InstallComponents: Reboot needed by windows update On Wed, 27 Feb 2008 10:38:16 +1100, Doron Holan <xxxxx@microsoft.com> wrote: > Which WDK are you using to build your driver? 6000 or 6001? What OS > are you trying to install on? It looks like Vista from your description. > > Is this a PNP driver? If so you cannot use sc.exe to start/stop it. > You must device manager or devcon to install the driver and then > enable/disable it to force load/unload. If this is a non pnp driver > (which is what I am guessing this is based on the samples you mention), > there is no INF to run which means the wdf coinstaller does not > automatically get into the mix. As such you need to manually invoke the > class installer to install the appropriate kmdf runtime. The nonpnp <...excess quoted lines suppressed...> -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/

Message 4 of 46

26 Feb 08 20:07

Tim Roberts

xxxxxx@probo.com

Join Date: 28 Jan 2005

Posts To This List: 8251

Re: Handle Invalid error when trying to run a KMDF driver

Joseph Thomas-Kerr wrote: > > I have now tried out nonpnp as well, and get what seem to be > equivalent symptoms. It asks me to reboot (I doubt this is intended), > but after I do and rerun nonpnpapp.exe it just asks me to reboot > again. It seems to be tripping over upgrading wdf01000.sys. I have > pasted the setupact.log below. > > Regards, > Joe. > <...excess quoted lines suppressed...> Yes, unfortunately, the reboot is intentional. This is the rather significant downside of the decision to ship KMDF as a kernel DLL instead of a linkable library, like USBCAMD. If any KMDF driver is loaded, installing another device with a newer KMDF will require a reboot. I suspect there are going to be some angry support calls about it, and it is only going to get worse as KMDF gets more popular. The hope, of course, is that KMDF will stabilize very quickly, and that 1.7 will last us many years. The confusion here seems to stem from the installer's determination that a KMDF driver is running ("Service Wdf01000 is running"), even though it couldn't contact it. It could be that there is a registry confusion. -- Tim Roberts, xxxxx@probo.com Providenza & Boekelheide, Inc.

Message 5 of 46

26 Feb 08 20:20

ntdev member 38016

xxxxxx@yahoo.com.au

Join Date:

Posts To This List: 23

Re: Handle Invalid error when trying to run a KMDF driver

Hi Tim, Thanks for the info. But its not the reboot that I am complaining about. Its the fact that KMDF doesn't work before or after the reboot. Regards, Joe. On Wed, 27 Feb 2008 12:06:24 +1100, Tim Roberts <xxxxx@probo.com> wrote: > Joseph Thomas-Kerr wrote: >> >> I have now tried out nonpnp as well, and get what seem to be equivalent >> symptoms. It asks me to reboot (I doubt this is intended), but after I >> do and rerun nonpnpapp.exe it just asks me to reboot again. It seems to >> be tripping over upgrading wdf01000.sys. I have pasted the setupact.log >> below. >> >> Regards, >> Joe. <...excess quoted lines suppressed...> -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/

Message 6 of 46

26 Feb 08 20:56

ntdev member 30611

xxxxxx@microsoft.com

Join Date:

Posts To This List: 132

RE: Handle Invalid error when trying to run a KMDF driver

We've been getting questions about problems with KMDF 1.7 installation on top of Vista RTM. There's an issue with a recent update (http://support.microsoft.com/kb/938371) released by MS as a prerequisite for Windows Vista Service Pack 1 (SP1). It is preventing installation of KMDF 1.7 due to versioning issues. We're investigating the causes of the version problems and are trying to come up with a solution. In the meantime, we suggest rejecting (or delaying) installation of that patch on your test machines. Also, we ask that you do not release driver packages with the currently available KMDF 1.7 co-installer until we've got a solution for this problem. Ilias -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Joseph Thomas-Kerr Sent: Tuesday, February 26, 2008 5:20 PM To: Windows System Software Devs Interest List Subject: Re: [ntdev] Handle Invalid error when trying to run a KMDF driver Hi Tim, Thanks for the info. But its not the reboot that I am complaining about. Its the fact that KMDF doesn't work before or after the reboot. Regards, Joe. On Wed, 27 Feb 2008 12:06:24 +1100, Tim Roberts <xxxxx@probo.com> wrote: > Joseph Thomas-Kerr wrote: >> >> I have now tried out nonpnp as well, and get what seem to be equivalent >> symptoms. It asks me to reboot (I doubt this is intended), but after I >> do and rerun nonpnpapp.exe it just asks me to reboot again. It seems to >> be tripping over upgrading wdf01000.sys. I have pasted the setupact.log >> below. >> >> Regards, >> Joe. <...excess quoted lines suppressed...> -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Message 7 of 46

26 Feb 08 21:26

ntdev member 30611

xxxxxx@microsoft.com

Join Date:

Posts To This List: 132

RE: Handle Invalid error when trying to run a KMDF driver

Also, another workaround that you can do for testing purposes in your test machines is to "disassemble" the coinstaller, as is described at http://blogs.msdn.com/bobkjelgaard/archive/2007/12/13/kmdf-1-7-and-the-server-200 8-vista-sp1-rc1-wdk.aspx, get the KMDF files (wdf0100.sys and wdfldr.sys) and use the technique described at http://kernelmustard.com/2007/03/27/the-death-of-wfp/ to install them in your Vista RTM test machines. As I said above, we're already working on providing a permanent solution for this issue. Ilias -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Ilias Tsigkogiannis Sent: Tuesday, February 26, 2008 5:55 PM To: Windows System Software Devs Interest List Subject: RE: [ntdev] Handle Invalid error when trying to run a KMDF driver We've been getting questions about problems with KMDF 1.7 installation on top of Vista RTM. There's an issue with a recent update (http://support.microsoft.com/kb/938371) released by MS as a prerequisite for Windows Vista Service Pack 1 (SP1). It is preventing installation of KMDF 1.7 due to versioning issues. We're investigating the causes of the version problems and are trying to come up with a solution. In the meantime, we suggest rejecting (or delaying) installation of that patch on your test machines. Also, we ask that you do not release driver packages with the currently available KMDF 1.7 co-installer until we've got a solution for this problem. Ilias -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Joseph Thomas-Kerr Sent: Tuesday, February 26, 2008 5:20 PM To: Windows System Software Devs Interest List Subject: Re: [ntdev] Handle Invalid error when trying to run a KMDF driver Hi Tim, Thanks for the info. But its not the reboot that I am complaining about. Its the fact that KMDF doesn't work before or after the reboot. Regards, Joe. On Wed, 27 Feb 2008 12:06:24 +1100, Tim Roberts <xxxxx@probo.com> wrote: > Joseph Thomas-Kerr wrote: >> >> I have now tried out nonpnp as well, and get what seem to be equivalent >> symptoms. It asks me to reboot (I doubt this is intended), but after I >> do and rerun nonpnpapp.exe it just asks me to reboot again. It seems to >> be tripping over upgrading wdf01000.sys. I have pasted the setupact.log >> below. >> >> Regards, >> Joe. <...excess quoted lines suppressed...> -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Message 8 of 46

27 Feb 08 01:15

ntdev member 38016

xxxxxx@yahoo.com.au

Join Date:

Posts To This List: 23

Re: Handle Invalid error when trying to run a KMDF driver

Thanks for the info. It's done the trick :). Apologies for the cross-posting. On Wed, 27 Feb 2008 10:25:39 +0800, Ilias Tsigkogiannis <xxxxx@microsoft.com> wrote: > Also, another workaround that you can do for testing purposes in your > test machines is to "disassemble" the coinstaller, as is described at > http://blogs.msdn.com/bobkjelgaard/archive/2007/12/13/kmdf-1-7-and-the-server-200 8-vista-sp1-rc1-wdk.aspx, > get the KMDF files (wdf0100.sys and wdfldr.sys) and use the technique > described at http://kernelmustard.com/2007/03/27/the-death-of-wfp/ to > install them in your Vista RTM test machines. > > As I said above, we're already working on providing a permanent solution > for this issue. > <...excess quoted lines suppressed...> -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/

Message 9 of 46

27 Feb 08 02:54

ntdev member 32323

xxxxxx@hotmail.com

Join Date:

Posts To This List: 843

Re: Handle Invalid error when trying to run a KMDF driver

Tim Roberts wrote: > The hope, of course, is that KMDF will stabilize very quickly, and that > 1.7 will last us many years. IIRC someone here stated that KMDF 1.9 is already in QA. Looks like this hope is in vain. :-)

Message 10 of 46

27 Feb 08 10:54

Doron Holan

xxxxxx@microsoft.com

Join Date: 08 Sep 2005

Posts To This List: 8270

RE: Re:Handle Invalid error when trying to run a KMDF driver

There will typically be a new version of WDF (both kmdf and umdf) in each OS release. This is not a sign of immaturity though. While each release brings bug fixes and new features (as requested by you our customers), the core functionality of the framework remains the same. If we did not do this, I am sure people would complain that we released a framework and then failed to continue to support, innovate, and keep it relevant for today's driver writers. d -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Hagen Patzke Sent: Tuesday, February 26, 2008 11:54 PM To: Windows System Software Devs Interest List Subject: Re:[ntdev] Handle Invalid error when trying to run a KMDF driver Tim Roberts wrote: > The hope, of course, is that KMDF will stabilize very quickly, and that > 1.7 will last us many years. IIRC someone here stated that KMDF 1.9 is already in QA. Looks like this hope is in vain. :-) --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Message 11 of 46

27 Feb 08 11:33

ntdev member 32323

xxxxxx@hotmail.com

Join Date:

Posts To This List: 843

Re: Handle Invalid error when trying to run a KMDF driver

Doron Holan wrote: > There will typically be a new version of WDF (both kmdf and umdf) in > each OS release. How about treating it as a "Windows System Component", then? Making it available as an extra download (similar to DirectX). It should also be automatically updated by Windows Update. Then it would not be necessary to always include a co-installer with a driver - only to state a requirement for a _minimum_ WDF version in the driver's INF file. Currently, for every WDF driver you always have to include the relevant co-installer. And in most cases the one delivered with the driver is probably already superseded by a more current version. Also newer versions of the framework completely _replace_ older ones. So why not handle it as an extra component and stop wasting all this install space and bandwith? OK, (now that MS started it) I would not take away the _option_ of delivering a WDF co-installer in a package with a driver. But I always wondered why there was not just a "WDF_install_X_yy_z.msi" available. And if a driver needs a newer version, I would get a message (something like "Required WDF version for this driver is v1.9. Currently installed version is v1.5. Upgrade now?"). -H

Message 12 of 46

27 Feb 08 14:41

ntdev member 8437

xxxxxx@windows.microsoft.com

Join Date:

Posts To This List: 1404

RE: Re:Handle Invalid error when trying to run a KMDF driver

It's a tempting idea ... the problem is that it then ties all this new device installation technology into specific versions of Windows. Sure your driver package could state that it needs update Y but the code to understand that is probably in update X which has to be applied first. The only way to resolve the chicken-and-egg problem on downlevel versions of Windows is to use the coinstaller. I'm hoping in the future we can make the coinstaller a little more general purpose since the problems facing KMDF are the same ones that face UMDF, WinUSB, and any other device technology we release out-of-band, but that's my hope not a plan or a promise. We've made the assumption that no one wants to have separate installation packages - one for pre Win7 and another for Win7 to deal with a radical change in the installation model. So we instead have to use a coinstaller. -p -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Hagen Patzke Sent: Wednesday, February 27, 2008 8:33 AM To: Windows System Software Devs Interest List Subject: Re:[ntdev] Handle Invalid error when trying to run a KMDF driver Doron Holan wrote: > There will typically be a new version of WDF (both kmdf and umdf) in > each OS release. How about treating it as a "Windows System Component", then? Making it available as an extra download (similar to DirectX). It should also be automatically updated by Windows Update. Then it would not be necessary to always include a co-installer with a driver - only to state a requirement for a _minimum_ WDF version in the driver's INF file. Currently, for every WDF driver you always have to include the relevant co-installer. And in most cases the one delivered with the driver is probably already superseded by a more current version. Also newer versions of the framework completely _replace_ older ones. So why not handle it as an extra component and stop wasting all this install space and bandwith? OK, (now that MS started it) I would not take away the _option_ of delivering a WDF co-installer in a package with a driver. But I always wondered why there was not just a "WDF_install_X_yy_z.msi" available. And if a driver needs a newer version, I would get a message (something like "Required WDF version for this driver is v1.9. Currently installed version is v1.5. Upgrade now?"). -H --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Message 13 of 46

27 Feb 08 14:49

Don Burn

xxxxxx@acm.org

Join Date:

Posts To This List: 3179

Re: Re:Handle Invalid error when trying to run a KMDF driver

One problem I don't see how you will address with the co-installer approach is the need for drivers to run at Windows installation time. I now have had two different customers who jumped on the KMDF bandwagon and then discovered after rewriting their drivers to KMDF that they could not use them as part of an install. Since in both cases the drivers were needed for Windows to work, both firms now have corporate edicts "NO WDF!!!!!" This is frustrating, since I have just finished some new drivers for them that otherwise would be perfect for KMDF, but need to be part of the install. -- Don Burn (MVP, Windows DDK) Windows 2k/XP/2k3 Filesystem and Driver Consulting Website: http://www.windrvr.com Blog: http://msmvps.com/blogs/WinDrvr Remove StopSpam to reply "Peter Wieland" <xxxxx@windows.microsoft.com> wrote in message news:102541@ntdev... It's a tempting idea ... the problem is that it then ties all this new device installation technology into specific versions of Windows. Sure your driver package could state that it needs update Y but the code to understand that is probably in update X which has to be applied first. The only way to resolve the chicken-and-egg problem on downlevel versions of Windows is to use the coinstaller. I'm hoping in the future we can make the coinstaller a little more general purpose since the problems facing KMDF are the same ones that face UMDF, WinUSB, and any other device technology we release out-of-band, but that's my hope not a plan or a promise. We've made the assumption that no one wants to have separate installation packages - one for pre Win7 and another for Win7 to deal with a radical change in the installation model. So we instead have to use a coinstaller. -p

Message 14 of 46

27 Feb 08 14:56

Tim Roberts

xxxxxx@probo.com

Join Date: 28 Jan 2005

Posts To This List: 8251

Re: Handle Invalid error when trying to run a KMDF driver

Peter Wieland wrote: > ... > The only way to resolve the chicken-and-egg problem on downlevel versions of Windows is to use the coinstaller. I'm hoping in the future we can make the coinstaller a little more general purpose since the problems facing KMDF are the same ones that face UMDF, WinUSB, and any other device technology we release out-of-band, but that's my hope not a plan or a promise. > > We've made the assumption that no one wants to have separate installation packages - one for pre Win7 and another for Win7 to deal with a radical change in the installation model. So we instead have to use a coinstaller. > Making it a statically linked library would have solved many of these issues. If my driver is tested with KMDF 1.5, then it can use KMDF forever, even if other drivers are using 1.7 and 1.9. May I ask why the decision was made to call the helper wdf01000.sys, instead of wdf01001.sys, wdf01005.sys, and wdf01007.sys? That, also, would have solved this problem, at the expense of 0.025c worth of disk space. -- Tim Roberts, xxxxx@probo.com Providenza & Boekelheide, Inc.

Message 15 of 46

27 Feb 08 15:12

ntdev member 8437

xxxxxx@windows.microsoft.com

Join Date:

Posts To This List: 1404

RE: Handle Invalid error when trying to run a KMDF driver

And would have resulted in a large body of MS kernel-mode code which couldn't be detected, serviced or blocked if a major security vulnerability was found. And while you might be kind enough to re-release your drivers should such an issue be found I'm quite certain there are many many other drivers that would never get re-released and would just continue to propagate that flaw across the entire ecosystem. Side-by-side support has a similar issue. From an IT administrator point of view it's better to have a single patch which will fix a problem and can be rolled back than to have a new form of DLL hell where each driver has to be managed individually. -p -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Tim Roberts Sent: Wednesday, February 27, 2008 11:56 AM To: Windows System Software Devs Interest List Subject: Re: [ntdev] Handle Invalid error when trying to run a KMDF driver Peter Wieland wrote: > ... > The only way to resolve the chicken-and-egg problem on downlevel versions of Windows is to use the coinstaller. I'm hoping in the future we can make the coinstaller a little more general purpose since the problems facing KMDF are the same ones that face UMDF, WinUSB, and any other device technology we release out-of-band, but that's my hope not a plan or a promise. > > We've made the assumption that no one wants to have separate installation packages - one for pre Win7 and another for Win7 to deal with a radical change in the installation model. So we instead have to use a coinstaller. > Making it a statically linked library would have solved many of these issues. If my driver is tested with KMDF 1.5, then it can use KMDF forever, even if other drivers are using 1.7 and 1.9. May I ask why the decision was made to call the helper wdf01000.sys, instead of wdf01001.sys, wdf01005.sys, and wdf01007.sys? That, also, would have solved this problem, at the expense of 0.025c worth of disk space. -- Tim Roberts, xxxxx@probo.com Providenza & Boekelheide, Inc. --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Message 16 of 46

27 Feb 08 15:28

ntdev member 30611

xxxxxx@microsoft.com

Join Date:

Posts To This List: 132

RE: Handle Invalid error when trying to run a KMDF driver

I'd like to give some more info on the issue of using an msi to update WDF. As you know, there are 2 ways to install a device and its driver: 1) Hardware-first: Plug-in the device, point the "add-new hardware wizard" to the inf and install the driver 2) Software-first: Install the driver, then plug-in the device In KMDF and UMDF we want to provide support for both ways and we don't want to change the model that windows drivers have been using so far. This means that we need an inf to install the driver. If we provided just an msi, which you'd have to install before pluging-in your device, this means that we're breaking the hardware-first installation. We don't want to do that, that's why we have the coinstaller So, the inf calls the coinstaller, which includes a windows update package. This package updates the WDF binaries that are on disk. From the user's perspective, it doesn't have any difference, if the coinstaller is internally using an msi, a windows update package or any other technology. So, just by replacing the update technology and using an msi, wouldn't give any benefit. On the contrary, we'd have to start development and testing from scratch and find solutions to problems like how it's possible to overcome Secure File Protection in Windows Vista in such a way that we don't break support for Windows 2000. At the same time, we also need to keep our binary small. That's why we're not changing the current architecture (yet). Having said that, I'd like to say that we've spent a considerable amount of time testing the KMDF 1.7 coinstaller and its behavior in key areas is very different than the KMDF 1.5 one. KB937381 was released after the testing of coinstallers was finished, so it was impossible to predict something like that. Even in this case, though, we were informed about the problem last week and within a few days we were able to find what's causing it and provide 2 workarounds. At the same time we're working on providing a permanent fix. Indeed, we understand the pain of driver installation, that's why many WDF developers and testers are monitoring ntdev, as well as other relevant mailing lists and newsgroups, so that we're always on speed with what problems people are facing. Ilias ________________________________________ From: xxxxx@lists.osr.com [xxxxx@lists.osr.com] On Behalf Of Peter Wieland [xxxxx@windows.microsoft.com] Sent: Wednesday, February 27, 2008 12:11 PM To: Windows System Software Devs Interest List Subject: RE: [ntdev] Handle Invalid error when trying to run a KMDF driver And would have resulted in a large body of MS kernel-mode code which couldn't be detected, serviced or blocked if a major security vulnerability was found. And while you might be kind enough to re-release your drivers should such an issue be found I'm quite certain there are many many other drivers that would never get re-released and would just continue to propagate that flaw across the entire ecosystem. Side-by-side support has a similar issue. From an IT administrator point of view it's better to have a single patch which will fix a problem and can be rolled back than to have a new form of DLL hell where each driver has to be managed individually. -p -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Tim Roberts Sent: Wednesday, February 27, 2008 11:56 AM To: Windows System Software Devs Interest List Subject: Re: [ntdev] Handle Invalid error when trying to run a KMDF driver Peter Wieland wrote: > ... > The only way to resolve the chicken-and-egg problem on downlevel versions of Windows is to use the coinstaller. I'm hoping in the future we can make the coinstaller a little more general purpose since the problems facing KMDF are the same ones that face UMDF, WinUSB, and any other device technology we release out-of-band, but that's my hope not a plan or a promise. > > We've made the assumption that no one wants to have separate installation packages - one for pre Win7 and another for Win7 to deal with a radical change in the installation model. So we instead have to use a coinstaller. > Making it a statically linked library would have solved many of these issues. If my driver is tested with KMDF 1.5, then it can use KMDF forever, even if other drivers are using 1.7 and 1.9. May I ask why the decision was made to call the helper wdf01000.sys, instead of wdf01001.sys, wdf01005.sys, and wdf01007.sys? That, also, would have solved this problem, at the expense of 0.025c worth of disk space. -- Tim Roberts, xxxxx@probo.com Providenza & Boekelheide, Inc. --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Message 17 of 46

27 Feb 08 15:40

ntdev member 32323

xxxxxx@hotmail.com

Join Date:

Posts To This List: 843

Re: Handle Invalid error when trying to run a KMDF driver

Peter Wieland wrote: > It's a tempting idea ... the problem is that it then ties all this > new device installation technology into specific versions of Windows. > Sure your driver package could state that it needs update Y but the > code to understand that is probably in update X which has to be > applied first. So you are telling me that in all of these Windows older driver installers there is absolutely no mechanism to inform anybody that a system component necessary for driver installation is missing? Not even with a text message? Come on, even then there could be a 'stub' that informs the user of the fact. As we had (still have?) in Windows EXE files that tell you "This program does not run in DOS mode.". And don't underestimate the amount of install hassle ordinary Windows users have become used to during all these years of "Windows Experience". YOu use an outdated version? OK, it will even be harder. Or someone has to get a system architect to MS. Quick. Swap her for someone from marketing - they are excellent and surely can spare one person. > We've made the assumption that no one wants to have separate > installation packages - one for pre Win7 and another for Win7 to deal > with a radical change in the installation model. So we instead have > to use a coinstaller. People have become used to have separate drivers for Win16, Win32s, Win95, Win98, Win98SE, WinME, Win2000, Win2003Server, WinXP and WinVista. Oh, and the 64 bit versions, of course. So used that some of them who happen to be driver developers even deny the possibility of writing one driver for everything from Win98 (with limitations) up to Vista. ;-) Great - I'm looking forward to a brave new world where every driver package will even bigger than in earlier times because now not only a bloated install program is delivered (to make sure the files are actually found), but additionally with every WDF driver comes a coinstaller. Well, the price of progress, I guess... Sorry for ranting, but this reminds me painfully of a three-disk installer for an external CD writer that would not run on NT4 with a "file not found" message. Luckily it was an old archive type that I managed to find an unpacker for. Turns out that more 2 and 3/4 disks were actually the (compressed!) install program. The real driver consisted of one INF and two SYS files, totalling less than 200kB. And of course someone had made a typo in the INF file, and obviously forgot to test it on NT4. The standard Windows device installer was absolutely happy taking the INF/SYS from a subdirectory on a single 3.5" disk. Now we will at least a single CD. Or DVD. Heck, now I know why BlueRay had to be invented. (Grrrrrr!) So much my 2cents of steam-off. No offense.

Message 18 of 46

27 Feb 08 15:47

ntdev member 19760

xxxxxx@evitechnology.com

Join Date:

Posts To This List: 2209

Re: Handle Invalid error when trying to run a KMDF driver

While were close to the subject, any idea of what a '.wdf' file with a size of 0 bytes that lives in '\Windows\System32\Drivers' does. For example, 'msftwdf_kernel_01005_coninstaller_critical.wdf.' They may be unrelated to WDF, they may be malware; I have no idea. I just saw three files like this 'msft_...' one one of my targets. Thanks, mm Tim Roberts wrote: > Peter Wieland wrote: >> ... >> The only way to resolve the chicken-and-egg problem on downlevel >> versions of Windows is to use the coinstaller. I'm hoping in the >> future we can make the coinstaller a little more general purpose since >> the problems facing KMDF are the same ones that face UMDF, WinUSB, and >> any other device technology we release out-of-band, but that's my hope >> not a plan or a promise. >> >> We've made the assumption that no one wants to have separate <...excess quoted lines suppressed...>

Message 19 of 46

27 Feb 08 15:51

ntdev member 30611

xxxxxx@microsoft.com

Join Date:

Posts To This List: 132

RE: Re:Handle Invalid error when trying to run a KMDF driver

The coinstaller just by itself is just a few kb. It's not bloated at all. The big size of the redistributable is caused by the update package that's inside the coinstaller. And yes, it's really easy to say "this driver cannot be installed in the system. Please use wdf17.msi". However, do think that customers would like it, if they are not able to do hardware-first device installations? For an additional explation, please look at my previous email. Ilias ________________________________________ From: xxxxx@lists.osr.com [xxxxx@lists.osr.com] On Behalf Of Hagen Patzke [xxxxx@hotmail.com] Sent: Wednesday, February 27, 2008 12:40 PM To: Windows System Software Devs Interest List Subject: Re:[ntdev] Handle Invalid error when trying to run a KMDF driver Peter Wieland wrote: > It's a tempting idea ... the problem is that it then ties all this > new device installation technology into specific versions of Windows. > Sure your driver package could state that it needs update Y but the > code to understand that is probably in update X which has to be > applied first. So you are telling me that in all of these Windows older driver installers there is absolutely no mechanism to inform anybody that a system component necessary for driver installation is missing? Not even with a text message? Come on, even then there could be a 'stub' that informs the user of the fact. As we had (still have?) in Windows EXE files that tell you "This program does not run in DOS mode.". And don't underestimate the amount of install hassle ordinary Windows users have become used to during all these years of "Windows Experience". YOu use an outdated version? OK, it will even be harder. Or someone has to get a system architect to MS. Quick. Swap her for someone from marketing - they are excellent and surely can spare one person. > We've made the assumption that no one wants to have separate > installation packages - one for pre Win7 and another for Win7 to deal > with a radical change in the installation model. So we instead have > to use a coinstaller. People have become used to have separate drivers for Win16, Win32s, Win95, Win98, Win98SE, WinME, Win2000, Win2003Server, WinXP and WinVista. Oh, and the 64 bit versions, of course. So used that some of them who happen to be driver developers even deny the possibility of writing one driver for everything from Win98 (with limitations) up to Vista. ;-) Great - I'm looking forward to a brave new world where every driver package will even bigger than in earlier times because now not only a bloated install program is delivered (to make sure the files are actually found), but additionally with every WDF driver comes a coinstaller. Well, the price of progress, I guess... Sorry for ranting, but this reminds me painfully of a three-disk installer for an external CD writer that would not run on NT4 with a "file not found" message. Luckily it was an old archive type that I managed to find an unpacker for. Turns out that more 2 and 3/4 disks were actually the (compressed!) install program. The real driver consisted of one INF and two SYS files, totalling less than 200kB. And of course someone had made a typo in the INF file, and obviously forgot to test it on NT4. The standard Windows device installer was absolutely happy taking the INF/SYS from a subdirectory on a single 3.5" disk. Now we will at least a single CD. Or DVD. Heck, now I know why BlueRay had to be invented. (Grrrrrr!) So much my 2cents of steam-off. No offense. --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Message 20 of 46

27 Feb 08 15:52

ntdev member 8437

xxxxxx@windows.microsoft.com

Join Date:

Posts To This List: 1404

RE: Re:Re:Handle Invalid error when trying to run a KMDF driver

You mean drivers that need to be loaded in txtsetup I assume? -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Don Burn Sent: Wednesday, February 27, 2008 11:49 AM To: Windows System Software Devs Interest List Subject: Re:[ntdev] Re:Handle Invalid error when trying to run a KMDF driver One problem I don't see how you will address with the co-installer approach is the need for drivers to run at Windows installation time. I now have had two different customers who jumped on the KMDF bandwagon and then discovered after rewriting their drivers to KMDF that they could not use them as part of an install. Since in both cases the drivers were needed for Windows to work, both firms now have corporate edicts "NO WDF!!!!!" This is frustrating, since I have just finished some new drivers for them that otherwise would be perfect for KMDF, but need to be part of the install. -- Don Burn (MVP, Windows DDK) Windows 2k/XP/2k3 Filesystem and Driver Consulting Website: http://www.windrvr.com Blog: http://msmvps.com/blogs/WinDrvr Remove StopSpam to reply "Peter Wieland" <xxxxx@windows.microsoft.com> wrote in message news:102541@ntdev... It's a tempting idea ... the problem is that it then ties all this new device installation technology into specific versions of Windows. Sure your driver package could state that it needs update Y but the code to understand that is probably in update X which has to be applied first. The only way to resolve the chicken-and-egg problem on downlevel versions of Windows is to use the coinstaller. I'm hoping in the future we can make the coinstaller a little more general purpose since the problems facing KMDF are the same ones that face UMDF, WinUSB, and any other device technology we release out-of-band, but that's my hope not a plan or a promise. We've made the assumption that no one wants to have separate installation packages - one for pre Win7 and another for Win7 to deal with a radical change in the installation model. So we instead have to use a coinstaller. -p --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Message 21 of 46

27 Feb 08 15:54

Don Burn

xxxxxx@acm.org

Join Date:

Posts To This List: 3179

Re: Re:Re:Handle Invalid error when trying to run a KMDF driver

Peter, Yes by txtsetup. This soured two firms that were going to convert almost everything to WDF, to saying that until there is a solution don't use the acronyms WDF, KMDF or UMDF in our presence. -- Don Burn (MVP, Windows DDK) Windows 2k/XP/2k3 Filesystem and Driver Consulting Website: http://www.windrvr.com Blog: http://msmvps.com/blogs/WinDrvr Remove StopSpam to reply "Peter Wieland" <xxxxx@windows.microsoft.com> wrote in message news:102552@ntdev... You mean drivers that need to be loaded in txtsetup I assume? -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Don Burn Sent: Wednesday, February 27, 2008 11:49 AM To: Windows System Software Devs Interest List Subject: Re:[ntdev] Re:Handle Invalid error when trying to run a KMDF driver One problem I don't see how you will address with the co-installer approach is the need for drivers to run at Windows installation time. I now have had two different customers who jumped on the KMDF bandwagon and then discovered after rewriting their drivers to KMDF that they could not use them as part of an install. Since in both cases the drivers were needed for Windows to work, both firms now have corporate edicts "NO WDF!!!!!" This is frustrating, since I have just finished some new drivers for them that otherwise would be perfect for KMDF, but need to be part of the install. -- Don Burn (MVP, Windows DDK) Windows 2k/XP/2k3 Filesystem and Driver Consulting Website: http://www.windrvr.com Blog: http://msmvps.com/blogs/WinDrvr Remove StopSpam to reply "Peter Wieland" <xxxxx@windows.microsoft.com> wrote in message news:102541@ntdev... It's a tempting idea ... the problem is that it then ties all this new device installation technology into specific versions of Windows. Sure your driver package could state that it needs update Y but the code to understand that is probably in update X which has to be applied first. The only way to resolve the chicken-and-egg problem on downlevel versions of Windows is to use the coinstaller. I'm hoping in the future we can make the coinstaller a little more general purpose since the problems facing KMDF are the same ones that face UMDF, WinUSB, and any other device technology we release out-of-band, but that's my hope not a plan or a promise. We've made the assumption that no one wants to have separate installation packages - one for pre Win7 and another for Win7 to deal with a radical change in the installation model. So we instead have to use a coinstaller. -p --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Message 22 of 46

27 Feb 08 15:59

Michal Vodicka

xxxxxx@upek.com

Join Date: 02 Apr 2004

Posts To This List: 1612

RE: Handle Invalid error when trying to run a KMDF driver

> -----Original Message----- > From: xxxxx@lists.osr.com > [mailto:xxxxx@lists.osr.com] On Behalf Of Peter Wieland > Sent: Wednesday, February 27, 2008 9:12 PM > To: Windows System Software Devs Interest List > Subject: RE: [ntdev] Handle Invalid error when trying to run > a KMDF driver > > And would have resulted in a large body of MS kernel-mode > code which couldn't be detected, serviced or blocked if a <...excess quoted lines suppressed...> It is the same as using static version of CRT library or anything else. If there is a vulnerability, it is the product which links this library what is vulnerable and has to be updated. I don't a real see difference between kernel and user mode. Vulnerable code can run within a service with sufficient privileges. Yes, it is easier for you to update framework binaries this way. It is also easier for you to non-intentionally break installed drivers using it. I don't say it is wrong; it is a tradeoff as usual. But there should be a choice. The same as between static and dynamic CRT. There are good reasons of both and it is producers' decision what reasons are more important for her. Best regards, Michal Vodicka UPEK, Inc. [xxxxx@upek.com, http://www.upek.com]

Message 23 of 46

27 Feb 08 16:01

ntdev member 8437

xxxxxx@windows.microsoft.com

Join Date:

Posts To This List: 1404

RE: Re:Handle Invalid error when trying to run a KMDF driver

We call them "marker files". The coinstaller creates them. When a crash is reported back to MS we get a listing of the drivers directory (but no actual file contents) and they get stored in the backend. It gives us a listing of the WDF drivers installed on that machine. -p -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Martin O'Brien Sent: Wednesday, February 27, 2008 12:47 PM To: Windows System Software Devs Interest List Subject: Re:[ntdev] Handle Invalid error when trying to run a KMDF driver While were close to the subject, any idea of what a '.wdf' file with a size of 0 bytes that lives in '\Windows\System32\Drivers' does. For example, 'msftwdf_kernel_01005_coninstaller_critical.wdf.' They may be unrelated to WDF, they may be malware; I have no idea. I just saw three files like this 'msft_...' one one of my targets. Thanks, mm Tim Roberts wrote: > Peter Wieland wrote: >> ... >> The only way to resolve the chicken-and-egg problem on downlevel >> versions of Windows is to use the coinstaller. I'm hoping in the >> future we can make the coinstaller a little more general purpose since >> the problems facing KMDF are the same ones that face UMDF, WinUSB, and >> any other device technology we release out-of-band, but that's my hope >> not a plan or a promise. >> >> We've made the assumption that no one wants to have separate <...excess quoted lines suppressed...> --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Message 24 of 46

27 Feb 08 16:14

ntdev member 19760

xxxxxx@evitechnology.com

Join Date:

Posts To This List: 2209

Re: Handle Invalid error when trying to run a KMDF driver

Thanks, Peter. mm Peter Wieland wrote: > We call them "marker files". The coinstaller creates them. When a crash is reported back to MS we get a listing of the drivers directory (but no actual file contents) and they get stored in the backend. It gives us a listing of the WDF drivers installed on that machine. > > -p > > -----Original Message----- > From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Martin O'Brien > Sent: Wednesday, February 27, 2008 12:47 PM > To: Windows System Software Devs Interest List > Subject: Re:[ntdev] Handle Invalid error when trying to run a KMDF driver > <...excess quoted lines suppressed...>

Message 25 of 46

27 Feb 08 16:30

ntdev member 8437

xxxxxx@windows.microsoft.com

Join Date:

Posts To This List: 1404

RE: Re:Handle Invalid error when trying to run a KMDF driver

As long as we're relying on magical things we've imagined that don't exist, we might as well have the computer call up and have a CD with WDF delivered to the user's home in a golden carriage pulled by badgers wearing little white gloves and top hats. Then the badgers can install it for you, give you a quick mink massage and go on their way without even leaving footprints. Unfortunately no such feature exists in previous versions of Windows (and I think the legal risks associated with the badger delivery service would be too high for Win7) so we're left with reality. The input we have gotten around setup was overwhelmingly that driver package had to be able to carry its dependencies with it. It couldn't rely on a service pack or WU update being on the machine. It couldn't rely on them coming down over the network. It has to work without any user interaction or UI. We have the resources to put together *one* setup method ... this was the most requested scenario and functions in the other cases and so it's the one we implemented. We have some ideas about how to streamline the installation but it's not clear when we'll be able to make those bubble up to the top of the list of all the things we're being asked to do (from external and internal partners). Don't worry about the rant. Clearly it was my mistake for trying to answer your question. My apologies for sending you off the deep end. -p -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Hagen Patzke Sent: Wednesday, February 27, 2008 12:40 PM To: Windows System Software Devs Interest List Subject: Re:[ntdev] Handle Invalid error when trying to run a KMDF driver Peter Wieland wrote: > It's a tempting idea ... the problem is that it then ties all this > new device installation technology into specific versions of Windows. > Sure your driver package could state that it needs update Y but the > code to understand that is probably in update X which has to be > applied first. So you are telling me that in all of these Windows older driver installers there is absolutely no mechanism to inform anybody that a system component necessary for driver installation is missing? Not even with a text message? Come on, even then there could be a 'stub' that informs the user of the fact. As we had (still have?) in Windows EXE files that tell you "This program does not run in DOS mode.". And don't underestimate the amount of install hassle ordinary Windows users have become used to during all these years of "Windows Experience". YOu use an outdated version? OK, it will even be harder. Or someone has to get a system architect to MS. Quick. Swap her for someone from marketing - they are excellent and surely can spare one person. > We've made the assumption that no one wants to have separate > installation packages - one for pre Win7 and another for Win7 to deal > with a radical change in the installation model. So we instead have > to use a coinstaller. People have become used to have separate drivers for Win16, Win32s, Win95, Win98, Win98SE, WinME, Win2000, Win2003Server, WinXP and WinVista. Oh, and the 64 bit versions, of course. So used that some of them who happen to be driver developers even deny the possibility of writing one driver for everything from Win98 (with limitations) up to Vista. ;-) Great - I'm looking forward to a brave new world where every driver package will even bigger than in earlier times because now not only a bloated install program is delivered (to make sure the files are actually found), but additionally with every WDF driver comes a coinstaller. Well, the price of progress, I guess... Sorry for ranting, but this reminds me painfully of a three-disk installer for an external CD writer that would not run on NT4 with a "file not found" message. Luckily it was an old archive type that I managed to find an unpacker for. Turns out that more 2 and 3/4 disks were actually the (compressed!) install program. The real driver consisted of one INF and two SYS files, totalling less than 200kB. And of course someone had made a typo in the INF file, and obviously forgot to test it on NT4. The standard Windows device installer was absolutely happy taking the INF/SYS from a subdirectory on a single 3.5" disk. Now we will at least a single CD. Or DVD. Heck, now I know why BlueRay had to be invented. (Grrrrrr!) So much my 2cents of steam-off. No offense. --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Message 26 of 46

27 Feb 08 16:59

ntdev member 32323

xxxxxx@hotmail.com

Join Date:

Posts To This List: 843

Re: Handle Invalid error when trying to run a KMDF driver

Peter Wieland wrote: > We've made the assumption that no one wants to have separate > installation packages - one for pre Win7 and another for Win7 to deal > with a radical change in the installation model. So we instead have > to use a coinstaller. So this model is "future-proof"? Wow. (Yes, only now it sank in that you were actually talking about Win7.) Well, I imagine WDF will be a huge success. In my opinion it is "condemned" to. (And that's why I get frightened by the perspective of installation collections with 134 drivers, where 105+ of them have outdated WDF co-installers, and 29 of them have the same, current, WDF co-installer in their package. Which in the end probably none gets actually installed at all, because the latest overnight auto-update installed the most current WDF anyway. Because it _is_ important.)

Message 27 of 46

27 Feb 08 17:02

ntdev member 32323

xxxxxx@hotmail.com

Join Date:

Posts To This List: 843

Re: Handle Invalid error when trying to run a KMDF driver

Peter Wieland wrote: > As long as we're relying on magical things we've imagined that don't exist, we might as well have the computer call up and have a CD with WDF delivered to the user's home in a golden carriage pulled by badgers wearing little white gloves and top hats. Then the badgers can install it for you, give you a quick mink massage and go on their way without even leaving footprints. Unfortunately no such feature exists in previous versions of Windows (and I think the legal risks associated with the badger delivery service would be too high for Win7) so we're left with reality. > > The input we have gotten around setup was overwhelmingly that driver package had to be able to carry its dependencies with it. It couldn't rely on a service pack or WU update being on the machine. It couldn't rely on them coming down over the network. It has to work without any user interaction or UI. > > We have the resources to put together *one* setup method ... this was the most requested scenario and functions in the other cases and so it's the one we implemented. We have some ideas about how to streamline the installation but it's not clear when we'll be able to make those bubble up to the top of the list of all the things we're being asked to do (from external and internal partners). > > Don't worry about the rant. Clearly it was my mistake for trying to answer your question. My apologies for sending you off the deep end. > > -p > <...excess quoted lines suppressed...>

Message 28 of 46

27 Feb 08 17:11

ntdev member 19760

xxxxxx@evitechnology.com

Join Date:

Posts To This List: 2209

Re: Handle Invalid error when trying to run a KMDF driver

Peter: This is not exactly hard to believe. I'm just not sure how much size matters here, as it were. That is, I can't really think of any reason why I would chose a smaller package (ibid) over being able to get one package and know that I was good to go, I would be mighty unhappy in many cases if a network connection were required, and finding out what you're missing only during install, unless you happen to know the contents of every machine on which you will be installing, while you read carefully the documentation for the driver installation, that must be accurate, is the worst case, for me, at least. Hagen: This is absurd: >People have become used to have separate drivers for Win16, Win32s, >Win95, Win98, Win98SE, WinME, Win2000, Win2003Server, WinXP and >WinVista. Oh, and the 64 bit versions, of course. >So used that some of them who happen to be driver developers even deny >the possibility of writing one driver for everything from Win98 (with >limitations) up to Vista. ;-) I gather you have the misfortune of supporting Win98, as you bring it up a lot, if I am remembering correctly. You're one of the few here that does. Perhaps it's possible that people merely don't talk about Win98, because they don't care. Also, of the ten versions of Windows you mentioned: 1. Win16 is pretty clearly not something has any expectations about either way at this point. 2. Win32s is something no one ever had any thoughts about at any point, and drivers really don't have much to do with it. 3. While there are some differences between all the Win95 family, not all that many people are concerned with any of these at this point, and I having trouble with the idea of anyone expecting a different driver for Win98 for Win98SE, at a minimum. 4. You left of WinNT 3.51 & 4.0. There's no reason to discuss them, as I can recall exactly one question on this list in three years about 4.0, but if you're going to talk about Win32s, why not these, and the Itanium for that matter as well. 4. Win2000 is still out there, but a lot of drivers that work on XP work in 2K as well. 5. Win2K3 supports almost everything that XP does. 6. XP is what the only thing that most people care about at the moment. 7. Vista is indeed different, and those who use it no doubt expect a separate driver package. 8. Also certainly true for 64-bit. What I really don't understand is given how outrageously awful installation has always been, easily the worst feature of windows, and that this is what causes a lot of issues today, and will continue to do so tomorrow, why one earth would focus on driver package size. How does this affect you? I mean, sure it takes longer to download, and if we were talking about why is the WDK like 2GB sometimes because of DTM, or that you have to download about 6 GB to create a partial checked build (unless you already have imagex), I would heartily agree with you, and add that the whole thing is made far worse by FTM, and that in some cases if you don't know to burn a lower speed, the OS iso's don't really work. But how big can a driver really be? Bigger than necessary, sure, but it doesn't even chart. mm Peter Wieland wrote: > As long as we're relying on magical things we've imagined that don't exist, we might as well have the computer call up and have a CD with WDF delivered to the user's home in a golden carriage pulled by badgers wearing little white gloves and top hats. Then the badgers can install it for you, give you a quick mink massage and go on their way without even leaving footprints. Unfortunately no such feature exists in previous versions of Windows (and I think the legal risks associated with the badger delivery service would be too high for Win7) so we're left with reality. > > The input we have gotten around setup was overwhelmingly that driver package had to be able to carry its dependencies with it. It couldn't rely on a service pack or WU update being on the machine. It couldn't rely on them coming down over the network. It has to work without any user interaction or UI. > > We have the resources to put together *one* setup method ... this was the most requested scenario and functions in the other cases and so it's the one we implemented. We have some ideas about how to streamline the installation but it's not clear when we'll be able to make those bubble up to the top of the list of all the things we're being asked to do (from external and internal partners). > > Don't worry about the rant. Clearly it was my mistake for trying to answer your question. My apologies for sending you off the deep end. > > -p > <...excess quoted lines suppressed...>

Message 29 of 46

27 Feb 08 17:24

ntdev member 32323

xxxxxx@hotmail.com

Join Date:

Posts To This List: 843

Re: Handle Invalid error when trying to run a KMDF driver

(Sorry about the empty quote before - I need to use "rewrap", don't have a shortcut for it, and accidentally clicked "send". :-( ) Peter Wieland wrote: > The input we have gotten around setup was overwhelmingly that driver > package had to be able to carry its dependencies with it. > It couldn't rely on a service pack or WU update being on the machine. > It couldn't rely on them coming down over the network. It has to > work without any user interaction or UI. OK. With this pretext the only solution is "carry everything with you". Fair enough. As Ilias already pointed out, the installer itself is actually very small, but then of course we do need the framework. And I also see that with the ability to install by device-plugin on a non-WDF-preinstalled system the WDF basically has to be in the package. It's of course a philosophy thing as well -- is something as WDF an important, "standalone" core system component? - see, its integrated in Vista! Then an installer for older systems can come on the same disk as the driver, but usually will be present with the latest SP on most systems. Analog to DirectX. Or like USB2.0 support for WinXP which IIRC needs SP2. Or is WDF "just" a helper component for UMDF/KMDF drivers that you basically bundle with your drivers, that's self-determining its most current version and that you then basically can forget about? Of course - at least for me it is obvious why - you do not want to go for statically linked libraries. And I also assume you don't want to have several versions of WDF to coexist on one machine (which is a good thing). > Don't worry about the rant. Clearly it was my mistake for trying to > answer your question. My apologies for sending you off the deep end. > I would very much regret if you feel it was a mistake to have tried answering my question. My apologies if you felt personally attacked - it is clearly not meant this way, and I do appreciate the chance to discuss these maters and to learn about the background and the "why" of things. (Of which thanks to your answer I'm doing quite a substantial amount in this thread!)

Message 30 of 46

27 Feb 08 17:39

ntdev member 30611

xxxxxx@microsoft.com

Join Date:

Posts To This List: 132

RE: Re:Handle Invalid error when trying to run a KMDF driver

Hagen, I didn't understand the point that you're trying to raise with this email. KMDF supports all operating systems from Windows 2000 to Windows Vista (both client and server editions). The latest auto-update that you mentioned exists only in Vista. How about the previous operating systems? We need to coinstaller to update Windows 2000 to the latest version of KMDF, because many driver developers are still writing drivers for Windows 2000. There is no latest auto-update for that OS (or for XP, etc). The same coinstaller makes sure that the framework will be installed in all supported OS versions. So, the update package has to be included. Also, it doesn't matter if a driver is shipped with an "outdated" (as you mention it) coinstaller. If the driver is installed in a system, which has a newer version of the framework (either UMDF or KMDF), then the driver automatically gets the benefits (stability, bug fixes, support, etc) that comes with the latest version. So, what exactly is the problem here? Ilias -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Hagen Patzke Sent: Wednesday, February 27, 2008 1:59 PM To: Windows System Software Devs Interest List Subject: Re:[ntdev] Handle Invalid error when trying to run a KMDF driver Peter Wieland wrote: > We've made the assumption that no one wants to have separate > installation packages - one for pre Win7 and another for Win7 to deal > with a radical change in the installation model. So we instead have > to use a coinstaller. So this model is "future-proof"? Wow. (Yes, only now it sank in that you were actually talking about Win7.) Well, I imagine WDF will be a huge success. In my opinion it is "condemned" to. (And that's why I get frightened by the perspective of installation collections with 134 drivers, where 105+ of them have outdated WDF co-installers, and 29 of them have the same, current, WDF co-installer in their package. Which in the end probably none gets actually installed at all, because the latest overnight auto-update installed the most current WDF anyway. Because it _is_ important.) --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Message 31 of 46

27 Feb 08 17:40

Tim Roberts

xxxxxx@probo.com

Join Date: 28 Jan 2005

Posts To This List: 8251

Re: Handle Invalid error when trying to run a KMDF driver

Martin O'Brien wrote: > > I gather you have the misfortune of supporting Win98, as you bring it > up a lot, if I am remembering correctly. You're one of the few here > that does. Perhaps it's possible that people merely don't talk about > Win98, because they don't care. Also, of the ten versions of Windows > you mentioned: > > 1. Win16 is pretty clearly not something has any expectations about > either way at this point. > <...excess quoted lines suppressed...> My experience is somewhat different. I do have clients that are still interested in Win98. I find that I need one driver package for 98/SE/ME, one for 2K/XP/2003/Vista32, and one for Vista64. For multimedia work, NT4 is kind of the poor stepchild, and never gets considered. I certainly do not expect to deliver separate binaries for XP and Vista32. -- Tim Roberts, xxxxx@probo.com Providenza & Boekelheide, Inc.

Message 32 of 46

27 Feb 08 17:52

ntdev member 32323

xxxxxx@hotmail.com

Join Date:

Posts To This List: 843

Re: Handle Invalid error when trying to run a KMDF driver

Martin O'Brien wrote: > I gather you have the misfortune of supporting Win98, as you bring it > up a lot, if I am remembering correctly. (a) Actually I am absolutely amazed that with WDM it is actually possible to make one driver package that supports a huge amount of Windows versions. It's impressive. (b) Win98 serves as an example that at some point you either have to "drop support" for something, or you will have to cater for it. If e.g. someone asks which technology to use, and everybody chimes in "WDF", IMO it just has to be stated what it can *not* be used for. With all the hot new technologies knocking at the doorstep it is also very easy to forget about maintenance of an existing customer base (or devices running a standard Windows but still "embedded" somewhere in a use case). In one year I'll move use WinME as an example instead, OK? Or should I go to NT4? Win2K would not hit the point as it *is* actually supported by WDF. ;-) > [Win versions] IIRC, KMDF is available from Win2000. UMDF from WinXP. WDF is included with Vista. So in my humble opinion one possible option to go for would have been to say "new drivers using WDF need some minimum WDF version to be installed". In my opinion people used to these old OS versions could be able to understand that in order to use a brand-new driver they needed an upgrade first. > What I really don't understand is given how outrageously awful > installation has always been, easily the worst feature of windows, > and that this is what causes a lot of issues today, and will continue > to do so tomorrow, why one earth would focus on driver package size. I am not concerned about the package size of the individual driver - if it is feature packaged, fine, ok, then make good use of your x megabyte package. But I am a bit sad about the prospect of having to bundle WDF for all new drivers to come, for years to com, and almost none of these installation packages will actually be used. Multiply this by the amount of archives, backups, etc. Some applications need DirectX Y.x or NET2.0 these days. With the very same argument everybody should as well bundle these inside their install packages. It is easier, and space doesn't matter anymore. The little extra wrapper on this bar of chocolate does not matter. But if you distribute billions of it there will be a considerable amount of waste. > But how big can a driver really be? Bigger than necessary, sure, but > it doesn't even chart. Ever tried to download a vital driver via a cell phone connection?

Message 33 of 46

27 Feb 08 17:52

ntdev member 19760

xxxxxx@evitechnology.com

Join Date:

Posts To This List: 2209

Re: Handle Invalid error when trying to run a KMDF driver

That anyone still supports much in the way of 95, et. c. is indeed not what I was expecting, and the new video driver model is the only reason I broke Vista out separately, but I suppose that is a very, very small group as well, that doesn't even really have to support it necessarily, so I see what you're saying, if I've understood you correctly. mm Tim Roberts wrote: > Martin O'Brien wrote: >> >> I gather you have the misfortune of supporting Win98, as you bring it >> up a lot, if I am remembering correctly. You're one of the few here >> that does. Perhaps it's possible that people merely don't talk about >> Win98, because they don't care. Also, of the ten versions of Windows >> you mentioned: >> >> 1. Win16 is pretty clearly not something has any expectations about >> either way at this point. <...excess quoted lines suppressed...>

Message 34 of 46

27 Feb 08 18:11

Tim Roberts

xxxxxx@probo.com

Join Date: 28 Jan 2005

Posts To This List: 8251

Re: Handle Invalid error when trying to run a KMDF driver

Hagen Patzke wrote: > > In my opinion people used to these old OS versions could be able to > understand that in order to use a brand-new driver they needed an > upgrade first. Apparently, you have never worked in a company that makes retail products. When you sell a commodity web camera, you simply do not have the luxury of dictating to your users that they must upgrade their operating system before installing it. Grandma and Grandpa don't even know that they HAVE an operating system. I can't tell you how many times I've asked "What operating system are you running?" only to be told "I'm running Office 2000." Further, you aren't going to want to be the one who stands in front of the vice president of marketing and tells her that she has to remove 98 from the list of operating systems in the marketing material. -- Tim Roberts, xxxxx@probo.com Providenza & Boekelheide, Inc.

Message 35 of 46

27 Feb 08 18:38

Peter Viscarola (OSR)

xxxxxx@osr.com

Join Date:

Posts To This List: 3926

List Moderator

RE: Handle Invalid error when trying to run a KMDF driver

top hats?? TOP HATS?? I'm all for those badgers, AND the little white gloves, but there's absolutely no need for top hats. They take up too much room, and they're inappropriate in Windows for Workgroups environments (where we all know, baseball caps must be worn. But that may have changed because it's a long time since I looked at it). I DO so enjoy Peter Wieland. In all seriousness, though... a statically linked option would be really helpful. I know the reason it wasn't chosen, but it would eliminate (a) the reboot problem, and (b) the legitimate (and not inconsiderable) fear in the community over running a driver with an up-rev version of the Framework with which the driver in question has never been tested. I, for one, would really hate to be the KMDF test lead that has to sign off that every driver that was working with KMDF V1.1 will work properly (out of the box, and without any changes) using KMDF V1.7 -- That responsibility would scare me to death. The reason it scares me so is that I can't see a way that I could even KNOW for certain it's true. I can know if it's NOT true, but... Peter OSR

Message 36 of 46

27 Feb 08 18:44

ntdev member 32323

xxxxxx@hotmail.com

Join Date:

Posts To This List: 843

Re: Handle Invalid error when trying to run a KMDF driver

Ilias Tsigkogiannis wrote: > KMDF supports all operating systems from Windows 2000 to Windows > Vista (both client and server editions). Yes, and IIRC, UMDF supports WinXP up to Vista. (Correct?) > The latest auto-update that you mentioned exists only in Vista. How > about the previous operating systems? AFAIK, Windows XP has got an update mechanism, too. At least the XP on my desktop computer keeps telling me regularly it wants to download and install something. True, certainly not as sophisticated as Vista. > Also, it doesn't matter if a driver is shipped with an "outdated" (as > you mention it) coinstaller. If the driver is installed in a system, > which has a newer version of the framework (either UMDF or KMDF), > then the driver automatically gets the benefits (stability, bug > fixes, support, etc) that comes with the latest version. Yes, you are right - from an in-system perspective, nothing bad will happen if a newly installed driver brings its own, "older" WDF version. From an outside-system perspective, this unused, bundled "older" WDF version is wasted space/bandwidth/time. > So, what exactly is the problem here? The law of big numbers: n WDF drivers bring n times a WDF install package with it, which in a high percentage of all install cases will not be installed. These n WDF drivers will be on product CDs, driver websites, magazine CDs, self-made "work" install CDs, on backups, tapes, harddisks, etc. Look at n for n >> 1000. Plus multiple driver versions (n >> 10000). As I perceive WDF (both KMDF and UMDF) as a very important and valuable system upgrade, I wonderend why it was not treated it as part of the system like printing, GDI, or like a central library like MFC (or one of the core system DLLs if you like). Of course "we decided to include WDF so it can automagically install pain-free on older systems" is a valid design decision! The question is if it is necessary and beneficial to bundle the framework with every single WDF driver - for years to come. IIRC, for USB2.0 support on WinXP, its SP2 needs to be installed. I wondered why not a similar setup was used, or a WDF installer for Win2000 and WinXP? And you install the latest "WDF support" when a driver needs it. Analogies are DirectX and .NET - they are not bundled with the install packages of the applications that use them. If you want to install an application that needs them, you're kindly informed what is missing. Even if their distributions are included on the product CD, you can skip these versions and go for a newer version (e.g. on a different CD, or you could even download the latest version).

Message 37 of 46

27 Feb 08 18:51

ntdev member 32323

xxxxxx@hotmail.com

Join Date:

Posts To This List: 843

Re: Handle Invalid error when trying to run a KMDF driver

> Hagen Patzke wrote: >> In my opinion people used to these old OS versions could be able to >> understand that in order to use a brand-new driver they needed an >> upgrade first. Tim Roberts wrote: > Apparently, you have never worked in a company that makes retail > products. No, not retail. Luckily. I think I _do_ understand. > Further, you aren't going to want to be the one who stands in front > of the vice president of marketing and tells her that she has to > remove 98 from the list of operating systems in the marketing > material. Sorry, this was bad wording: not an OS upgrade, but e.g. a "WDF upgrade package", like you need DirectX to run some games, or NET 2.0 for some applications, or (IIRC) SP2 for XP for USB2.0 support. On older systems you very often need an installer anyway, and IMHO that installer could also run a WDF pre-installation from the same product CD or driver package.

Message 38 of 46

27 Feb 08 18:51

ntdev member 30611

xxxxxx@microsoft.com

Join Date:

Posts To This List: 132

RE: Re:Handle Invalid error when trying to run a KMDF driver

-----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Hagen Patzke Sent: Wednesday, February 27, 2008 3:44 PM To: Windows System Software Devs Interest List Subject: Re:[ntdev] Handle Invalid error when trying to run a KMDF driver Analogies are DirectX and .NET - they are not bundled with the install packages of the applications that use them. If you want to install an application that needs them, you're kindly informed what is missing. Even if their distributions are included on the product CD, you can skip these versions and go for a newer version (e.g. on a different CD, or you could even download the latest version). [Ilias] As you said.. they are APPLICATIONS! We are talking about DRIVERS! There is a difference between these 2. Let's say that somebody buys a new device (keyboard/mouse/usb stick... it doesn't matter), plugs it in and gets a message saying "sorry, however Windows doesn't support your keyboard. Please go to www.microsoft.com/foo/bar, download our KMDF installer and try again". How happy do you think that this person would be? Especially now that he has to use his computer with a non-working keyboard/mouse? Ilias

Message 39 of 46

27 Feb 08 19:04

ntdev member 32323

xxxxxx@hotmail.com

Join Date:

Posts To This List: 843

Re: Handle Invalid error when trying to run a KMDF driver

Tim Roberts wrote: > My experience is somewhat different. I do have clients that are still > interested in Win98. I find that I need one driver package for > 98/SE/ME, one for 2K/XP/2003/Vista32, and one for Vista64. For > multimedia work, NT4 is kind of the poor stepchild, and never gets > considered. Reminds me of the fact that my Dad still has WinME on his computer. Now he is interested in one of the new flat rate offers for national phone and internet access, and I discussed with him that he will want a faster computer with a more recent OS once he's got hi-speed internet access. Probably I'll make sure he gets Vista32. If the "projected average use time" is eight years it has to be the most recent OS we can get now. :-)

Message 40 of 46

27 Feb 08 19:05

Michal Vodicka

xxxxxx@upek.com

Join Date: 02 Apr 2004

Posts To This List: 1612

RE: Handle Invalid error when trying to run a KMDF driver

> -----Original Message----- > From: xxxxx@lists.osr.com > [mailto:xxxxx@lists.osr.com] On Behalf Of Tim Roberts > Sent: Wednesday, February 27, 2008 11:40 PM > To: Windows System Software Devs Interest List > Subject: Re: [ntdev] Handle Invalid error when trying to run > a KMDF driver > > I certainly do not expect to deliver separate binaries for XP > and Vista32. We have separate install package per every OS version including XP and Vista32. The reason is simple: driver signing. I'm not sure if it is possible to have one .cat file for more OSes but even if it is, we don't want it. We'd have to always sign for all OSes at once. It is painful and time consuming process and sometimes we need a new signature for one OS quickly. In turn, our install packages for XP and Vista32 can contain different versions of the same driver. Best regards, Michal Vodicka UPEK, Inc. [xxxxx@upek.com, http://www.upek.com]

Message 41 of 46

27 Feb 08 19:34

ntdev member 32323

xxxxxx@hotmail.com

Join Date:

Posts To This List: 843

Re: Handle Invalid error when trying to run a KMDF driver

Ilias Tsigkogiannis wrote: > As you said.. they are APPLICATIONS! We are talking about DRIVERS! > There is a difference between these 2. Let's say that somebody buys a > new device (keyboard/mouse/usb stick... it doesn't matter), plugs it > in and gets a message saying "sorry, however Windows doesn't support > your keyboard. Please go to www.microsoft.com/foo/bar, download our > KMDF installer and try again". > > How happy do you think that this person would be? Especially now that > he has to use his computer with a non-working keyboard/mouse? (a) I know exactly how you feel in this case - I experienced it. For driver testing (also performance), I had to set up Win98SE on a system with USB keyboard and mouse. "Please click 'continue' to set up <something>", otherwise installation would not continue. The message came _before_ the USB drivers were active. (Yes, I resigned and got PS/2 devices at that point.) (b) Because of this I would never remove the OPTION to include WDF with a driver that needs it. And with core system components of course it DOES make sense to include it! But probably I don't need another WDF installer for my mobile phone firmware update driver, the web cam, the joypad driver, the bluetooth adapter, the scanner, the video grabber, the smart card reader, the printer and for all the other USB gizmos and gadgets that are on the market. For these I would advocate (1) a way of getting a message during WDF driver install, if the necessary WDF version is not yet installed, and (2) have a clean and lean install package that just installs the framework. Then I would go and grab the one with the highest number on it (=latest version, either from one of the product CDs or from the 'net), install it, *then* do all of the WDF driver installations, and be done. IIRC with less install time, fewer reboots (and fewer backup copies of older versions of the WDF in the bowels of the system) than if every one of these drivers brings its own framework version with it. Correct? Never mind - if I see this correctly, it should not be difficult to make something like a "pure" WDF framework install package. Given the fact that a lot of install packages contain anyway slack stuff (drivers for different OSes, "useful" applications, etc.) most probably the additional megabytes for the framework installers does probably really not matter much. I'll just have to get used to it.

Message 42 of 46

27 Feb 08 20:01

ntdev member 32323

xxxxxx@hotmail.com

Join Date:

Posts To This List: 843

Re: Handle Invalid error when trying to run a KMDF driver

Ilias Tsigkogiannis wrote: > I'd like to give some more info on the issue of using an msi to update WDF. > [...] Thanks for the background!

Message 43 of 46

27 Feb 08 20:10

ntdev member 30611

xxxxxx@microsoft.com

Join Date:

Posts To This List: 132

RE: Re:Handle Invalid error when trying to run a KMDF driver

What if the user doesn't have a network access? Or if he just installed Windows and his network card driver depends on a newer KMDF version? So, the driver that YOU created won't be working! And then he'll be blaming Microsoft... Sorry, that's a bad user experience. Better safe than sorry. We want our users to have good user experiences. Just plug-in the device, point to the inf and have it working. We don't want to tell a user who doesn't have an internet connection (for any reason) that "he needs to download this package from the internet, otherwise his device won't work". We've given lots of thought on all these issues and we have some options, however keep in mind that all sorts of people use WDF drivers and we don't want to ruin the user experience for any of them. Ilias -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Hagen Patzke Sent: Wednesday, February 27, 2008 4:34 PM To: Windows System Software Devs Interest List Subject: Re:[ntdev] Handle Invalid error when trying to run a KMDF driver Ilias Tsigkogiannis wrote: > As you said.. they are APPLICATIONS! We are talking about DRIVERS! > There is a difference between these 2. Let's say that somebody buys a > new device (keyboard/mouse/usb stick... it doesn't matter), plugs it > in and gets a message saying "sorry, however Windows doesn't support > your keyboard. Please go to www.microsoft.com/foo/bar, download our > KMDF installer and try again". > > How happy do you think that this person would be? Especially now that > he has to use his computer with a non-working keyboard/mouse? (a) I know exactly how you feel in this case - I experienced it. For driver testing (also performance), I had to set up Win98SE on a system with USB keyboard and mouse. "Please click 'continue' to set up <something>", otherwise installation would not continue. The message came _before_ the USB drivers were active. (Yes, I resigned and got PS/2 devices at that point.) (b) Because of this I would never remove the OPTION to include WDF with a driver that needs it. And with core system components of course it DOES make sense to include it! But probably I don't need another WDF installer for my mobile phone firmware update driver, the web cam, the joypad driver, the bluetooth adapter, the scanner, the video grabber, the smart card reader, the printer and for all the other USB gizmos and gadgets that are on the market. For these I would advocate (1) a way of getting a message during WDF driver install, if the necessary WDF version is not yet installed, and (2) have a clean and lean install package that just installs the framework. Then I would go and grab the one with the highest number on it (=latest version, either from one of the product CDs or from the 'net), install it, *then* do all of the WDF driver installations, and be done. IIRC with less install time, fewer reboots (and fewer backup copies of older versions of the WDF in the bowels of the system) than if every one of these drivers brings its own framework version with it. Correct? Never mind - if I see this correctly, it should not be difficult to make something like a "pure" WDF framework install package. Given the fact that a lot of install packages contain anyway slack stuff (drivers for different OSes, "useful" applications, etc.) most probably the additional megabytes for the framework installers does probably really not matter much. I'll just have to get used to it. --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Message 44 of 46

28 Feb 08 08:46

Maxim S. Shatskih

xxxxxx@storagecraft.com

Join Date: 20 Feb 2003

Posts To This List: 8628

Re: Handle Invalid error when trying to run a KMDF driver

>test platform is vista. It is not a pnp driver; it is a virtual device >that looks like an audio device to the user but actually uses winsock >kernel to send the audio elsewhere. Maybe it is a good idea to implement it as a user-mode DLL driver which is accessible via waveXxx functions? -- Maxim Shatskih, Windows DDK MVP StorageCraft Corporation xxxxx@storagecraft.com http://www.storagecraft.com

Message 45 of 46

28 Feb 08 09:00

Maxim S. Shatskih

xxxxxx@storagecraft.com

Join Date: 20 Feb 2003

Posts To This List: 8628

Re: Handle Invalid error when trying to run a KMDF driver

>Vista32. The reason is simple: driver signing. I'm not sure if it is >possible to have one .cat file for more OSes but even if it is, we don't Vista64 signature embedded to the .SYS file is fine on 2008 x64. -- Maxim Shatskih, Windows DDK MVP StorageCraft Corporation xxxxx@storagecraft.com http://www.storagecraft.com

Message 46 of 46

04 Mar 08 14:01

Doron Holan

xxxxxx@microsoft.com

Join Date: 08 Sep 2005

Posts To This List: 8270

RE: Re:Re:Re:Handle Invalid error when trying to run a KMDF driver

We did work in 1.7 to help with txt mode setup. You still have to extract the runtime binaries out of the coinstaller since the coinstaller is not a part of F5 installation. BTW, txtmode setup support was explicitly not a goal with KMDF when we started b/c we were not targeting storage (and all of its various requirements like guaranteed fwd progress) and we considered it too niche to impact the overall design....but we were definitely surprised when folks started using KMDF as a bus driver to enumerate a storage part and not necessarily to be a part of the storage stack itself. d -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Don Burn Sent: Wednesday, February 27, 2008 12:55 PM To: Windows System Software Devs Interest List Subject: Re:[ntdev] Re:Re:Handle Invalid error when trying to run a KMDF driver Peter, Yes by txtsetup. This soured two firms that were going to convert almost everything to WDF, to saying that until there is a solution don't use the acronyms WDF, KMDF or UMDF in our presence. -- Don Burn (MVP, Windows DDK) Windows 2k/XP/2k3 Filesystem and Driver Consulting Website: http://www.windrvr.com Blog: http://msmvps.com/blogs/WinDrvr Remove StopSpam to reply "Peter Wieland" <xxxxx@windows.microsoft.com> wrote in message news:102552@ntdev... You mean drivers that need to be loaded in txtsetup I assume? -----Original Message----- From: xxxxx@lists.osr.com [mailto:xxxxx@lists.osr.com] On Behalf Of Don Burn Sent: Wednesday, February 27, 2008 11:49 AM To: Windows System Software Devs Interest List Subject: Re:[ntdev] Re:Handle Invalid error when trying to run a KMDF driver One problem I don't see how you will address with the co-installer approach is the need for drivers to run at Windows installation time. I now have had two different customers who jumped on the KMDF bandwagon and then discovered after rewriting their drivers to KMDF that they could not use them as part of an install. Since in both cases the drivers were needed for Windows to work, both firms now have corporate edicts "NO WDF!!!!!" This is frustrating, since I have just finished some new drivers for them that otherwise would be perfect for KMDF, but need to be part of the install. -- Don Burn (MVP, Windows DDK) Windows 2k/XP/2k3 Filesystem and Driver Consulting Website: http://www.windrvr.com Blog: http://msmvps.com/blogs/WinDrvr Remove StopSpam to reply "Peter Wieland" <xxxxx@windows.microsoft.com> wrote in message news:102541@ntdev... It's a tempting idea ... the problem is that it then ties all this new device installation technology into specific versions of Windows. Sure your driver package could state that it needs update Y but the code to understand that is probably in update X which has to be applied first. The only way to resolve the chicken-and-egg problem on downlevel versions of Windows is to use the coinstaller. I'm hoping in the future we can make the coinstaller a little more general purpose since the problems facing KMDF are the same ones that face UMDF, WinUSB, and any other device technology we release out-of-band, but that's my hope not a plan or a promise. We've made the assumption that no one wants to have separate installation packages - one for pre Win7 and another for Win7 to deal with a radical change in the installation model. So we instead have to use a coinstaller. -p --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer --- NTDEV is sponsored by OSR For our schedule of WDF, WDM, debugging and other seminars visit: http://www.osr.com/seminars To unsubscribe, visit the List Server section of OSR Online at http://www.osronline.com/page.cfm?name=ListServer

Posting Rules
You may not post new threads You may not post replies You may not post attachments	You must login to OSR Online AND be a member of the ntdev list to be able to post.

All times are GMT -5. The time now is 11:58.

Contact Us - Osr Online Homepage - Top