Enhance td-shim/ResetVector to support test on non-TDX hardware platfroms

[jiangliu]

It help ease the development if we could test td-shim on non-TDX hardware platforms. Of course, all TDX related functionality will be disabled on non-TDX hardware platform.
Basically we need to enhance ResetVector to support boot from real mode and then switch protected as what expected by TDX.

[jyao1]

Agree from high level.

Question: What do you think the interface between hypervisor in non-tdx shim ?
Option 1: We reuse the normal non-tdx interface. E.g. Shim code runs from real mode. Shim gets memory info from CMOS.
Option 2: We reuse the td-shim interface. E.g. Shim code runs from protected mode. Shim gets the memory info via hob in RCX.

[jiangliu]

I like option2, we should share as much code as possible with tdx.
Actually we need a shim for td-shim to setup the environment defined/expected by tdx

[ariel-adam]

@jiangliu is this issue still relevant or can be closed?
If it's still relevant to what release do you think we should map it to (mid-November, end-December, mid-February etc...)?

[jinankjain]

I am looking to familiarize myself with td-shim. This sounds like a good introductory task. So I have a few question here if someone can help me answer that would be great. What about the special registers provided TDX environment. For example: I think there is a register to obtain the VP count. How would that be emulated in the non-TDX hardware?