fix: clarify prereq, installed, post-inst - #480
Conversation
Signed-off-by: Mike McKiernan <mmckiernan@nvidia.com>
Documentation preview |
| * - AMD Genoa / Milan | ||
| - AMD SEV-SNP |
There was a problem hiding this comment.
@manuelh-dev , I think you proposed removing this information or moving it. I took a look at the page and my instinct is to keep it:
- It's not a disaster as it is. (Yay!)
- I suspect the CPUs and technology are not yet widely prevalent and highlighting the required CPU features seems important. (For now.)
- I suspect it meets expectations--that there would be a matrix for CPUs.
There was a problem hiding this comment.
@mikemckiernan - my suggestion was to move the Host Kernel Version and Host Operating System fields, specifically, down into the list below - but after your re-organization below, I think we are good. Agree that we can keep this information here
mikemckiernan
left a comment
mikemckiernan
left a comment
There was a problem hiding this comment.
More of a question.
| - Version | ||
| - When It Is Needed | ||
| - How It Is Provided | ||
| * - `Key Broker Service (KBS) protocol <https://confidentialcontainers.org/docs/attestation/>`__ |
There was a problem hiding this comment.
@fitzthum @fidencio - I think we need a bit of a refinement here. I think the KBS protocol is installed implicitly by shipping the attestation agent, CDH in our guests. So, I think here we want to say that trustee is not installed - and provide a specific trustee version to install, instead.
Then, in above listing for Kata Containers we can say that this comes with a guest using the 0.4.0 KBS protocol. This may make it easier here in turn because we can say to install a trustee version using that same protocol version (or, a protocol version that is compatible, if a patch version increment of the protocol version means that both are still interoperable).
I really like this, thank you for these changes! I left one comment for discussion. |
| .. flat-table:: | ||
| :header-rows: 1 | ||
|
|
||
| * - Interface or Component |
There was a problem hiding this comment.
@mikemckiernan - if or as we make this change, we could also list genpolicy - we should then merge #479 after this one.
Signed-off-by: Mike McKiernan <mmckiernan@nvidia.com>
2 participants
No description provided.