Re: [PATCH v5 5/6] x86/tdx: Add a helper to query maximum Quote size

[Peter Fang]

On Tue, Sep 29, 2026 at 07:18:33AM -0700, Dave Hansen wrote:
> >>
> >> If the patch is suggested by Xu Uilun you can use Suggested-by:
> > 
> > I think tip's preference is to use some prose rather than a tag for
> > these things [1]. So I'll probably keep the prose unless you have
> > objections?
> > 
> > [1] https://lore.kernel.org/all/c1c9bddf-1c00-4625-a915-d545acd67fd0@xxxxxxxxx/
> 
> Actually, I'm not a huge fan of the prose in this case either.
> "Suggesting this" is not specific. What's the point of even mentioning it?
> 
> Responding to a minor review comment does not need to get documented,
> either in prose or in a tag. What part of the suggestion was so
> important that it needs to be remembered for all time as coming from Xu
> Yilun in the commit log?
> 
> It doesn't seem that important because I don't even see Xu Yilun on cc. ;)

Yeah, I guess I've just been keeping that line every version. It was
basically documenting an early review comment about tdg_vm_rd() being
able to read random fields. I can drop it.

> >>
> >> I think -ENODEV or -EOPNOTSUPP is more appropriate return value
> >> for unsupported case.
> > 
> > Hmm, I don't see either one being a common pattern for guest-side TDX
> > code. -ENODEV is usually used for some basic operation failure.
> > -EOPNOTSUPP is not used in the guest. But between the two I think
> > -EOPNOTSUPP probably fits better. Thanks.
> 
> Just leave it alone. All that matters is that it is non-zero. If it were
> going out to userspace, we could argue all day about it, but it's not.

I'll leave it as is. Thanks.