Re: [PATCH RFC 00/20] attestation: Add a common subsystem and TDX support

[Mimi Zohar]

On Thu, 2026-10-22 at 15:52 -0400, James Bottomley wrote:
> On Thu, 2026-10-22 at 16:45 -0300, Jason Gunthorpe wrote:
> > On Thu, Oct 22, 2026 at 03:37:46PM -0400, Mimi Zohar wrote:
> > 
> > > > Advisory entries shouldn't matter with a scheme like this.
> > > 
> > > IMA has something called violations (e.g. open-writers, ToMToU),
> > > which are more
> > > than just advisory.  They intentionally invalidate the IMA
> > > measurement list
> > > since the measurement cannot be trusted.
> > 
> > Maybe I got lost, but when James said "advisory" I understood this as
> > the TCG type that does not get included in a PCR extend.
> 
> No, I meant the TCG type where the event description isn't part of the
> measurement (and thus can contain pretty much anything).
> 
Thanks for the clarification.

> In IMA
> measurements, the path is part of the log but isn't part of the
> measurement (primarily because it's not reliable: a mount namespace
> could make it anything it wants).

You have a good memory!  Prior to IMA being upstreamed, only the file data hash
was extended into the TPM PCR.

The upstreamed 'ima' template extended the template data hash, containing the
file data hash and filename limited to 255 bytes.  Depending on the template
format (e.g. ima-ng, ima-ngv2, ima_sig, ima_sigv2), the file template hash
contains the full file pathname.

Mimi