Re: Win11 22H2 and CoInstallers

[Paul Durrant <xadimgnik@xxxxxxxxx>]

On 08/08/2023 21:40, Chuck Zmudzinski wrote:
> On 28/06/2023 15:33, Owen Smith wrote:
> > Looks like the HLK tests for Win11 22H2 fail if you specify a coinstaller.
> > I've not dug through the full implications of this yet, but looks like the
> > coinstallers will need removing going forwards.
> Hello,
>
> I am glad to see there is an effort to get these drivers compatible
> with Windows 11!
>
> Here is a message in a long thread seven years old that explains what
> happened to coinstallers:
>
> https://community.osr.com/discussion/comment/276218/#Comment_276218
>
> Quoting from that message:
>
> > 1.11 was the last version that used a coinstaller and went down level. Every
> > version afterward is only available on the os that released that version.
> Since there are still coinstallers in use in the PV drivers, the PV drivers
> must be using a version <= 1.11. After reading the thread linked to above,
> it appears the implications are that if the Windows PV drivers can be
> written for KMDF 1.11 (Windows 8) then the implication is that going forward,
> the earliest version that can be supported going forward is Windows 8 without
> a coinstaller.
The PV drivers are WDM. They make no use of KMDF.

> If the PV drivers are compatible with an even earlier version,
> then the earliest version that can be supported would be whatever that
> earlier version is. This means that it is likely it will be more difficult
> to support older and no longer supported versions of Windows and Windows 11
> with the same version of the drivers.
>
> Of course, unless I am mistaken, in addition to removing coinstallers, it will
> also be necessary to write a xentpm driver so Windows 11 can access the PV TPM
> device, and upgrade the TPM device to version 2.0 if that is not done yet.
If the TPM is emulated there should be no need to write a PV driver :-)

  Paul