Re: [PATCH v3 34/40] KVM: x86: Move misc "VALID MASK" defines from kvm_host.h => x86.c

[Huang, Kai]

On Fri, 2026-06-05 at 13:32 -0700, Sean Christopherson wrote:
> On Fri, Jun 05, 2026, Kai Huang wrote:
> > On Fri, 2026-05-29 at 15:22 -0700, Sean Christopherson wrote:
> > > Move a variety of "VALID MASK" defines, e.g. that capture which flags in
> > > a given ioctl are supported by KVM, from kvm_host.h to x86.c.  The set of
> > > valid flags/bits is very much a KVM-internal detail, as the hardcoded
> > > defines are often massaged at runtime, i.e. *directly* using the macros
> > > outside of KVM x86 would be actively dangerous.
> > 
> > I am bit confused about what do you mean by "hardcoded defines are ofen 
> > massaged at runtime".  Do you mean the supported flags can be changed at runtime
> > via ioctl?  But if we are talking about ioctl, then the flags must be part of
> > uABI, which shouldn't change?
> 
> Not by ioctl, but by KVM's setup code.  Is this more clear?
> 
>   Move a variety of "VALID MASK" defines, e.g. that capture which flags in
>   a given ioctl are supported by KVM, from kvm_host.h to x86.c.  The set of
>   valid flags/bits is very much a KVM-internal detail, as the values from the
>   hardcoded #defines are often captured and massaged by KVM's setup code, i.e.
>   *directly* using the macros outside of KVM x86 would be actively dangerous.

Yes it's clearer to me.  Thanks!