On Wed, 31 Mar 2021 08:44:23 +0200 Boris Petkov wrote: > On March 31, 2021 3:10:32 AM GMT+02:00, Kai Huang <kai.hu...@intel.com> > wrote: > > > The admin will be aware of > >such EPC > >allocation disjoint situation, and deploy host enclaves/KVM SGX guests > >accordingly. > > The admin will be aware because... > > 1) he's following our discussion? > > 2) he'll read the commit messages and hopefully understand? > > 3) we *actually* have documentation somewhere explaining how we envision that > stuff to be used? > > Or none of the above and he'll end up doing whatever and then he'll > eventually figure out that we don't support that use case but he's doing it > already anyway and we don't break userspace so we have to support it now and > we're stuck somewhere between a rock and a hard place? > > Hmm, I think we have enough misguided use cases as it is - don't need another > one.
How about adding explanation to Documentation/x86/sgx.rst?