Re: [PATCH 8/9] x86/virt/tdx: Exclude memory region hole within CMR as TDMR's reserved area

[Binbin Wu]

On 6/19/2024 9:23 AM, Huang, Kai wrote:
[...]
> > furthermore the alignement checks
> > suggest it's actually some sanity checking function. Furthermore if we
> > have:"
> >
> > ORDINARY_CMR,EMPTY_CMR,ORDINARY_CMR
> >
> > (Is such a scenario even possible), in this case we'll ommit also the
> > last ordinary cmr region?
> It cannot happen.
>
> The fact is:
>
> 1) CMR base/size are 4KB aligned.  This is architectural behaviour.
> 2) TDX architecturally supports 32 CMRs maximumly;
Do you think it's worth a comment to the definition of TDX_MAX_CMRS that 
the number is architectural?

> 3) In practice, TDX can just report the 'NUM_CMRS' metadata field as 32,
> but there can be empty/null CMRs following valid CMRs.
> 4) A empty/null CMR between valid CMRs cannot happen.