On Wed, Apr 24, 2024, Rick P Edgecombe wrote: > Long term though, I have been wondering about how to prevent TDX regressions > especially on the MMU pieces. It is one thing to have the TDX setups available > for maintainers, but most normal developers will likely not have access to TDX > HW for a bit. Just a problem without a solution. I wouldn't worry too much about hardware availability. As you said, it's not a problem we can really solve, and we already have to be concious of the fact that not all developers have comparable hardware. E.g. most people don't have a 4-sock, multi-hundred CPU system with TiBs of RAM. Not being able to test at all is obviously a little different, but it's not entirely new. Instead, I would encourage spending time and effort (after things have settled down patch wise) to build out selftests. I tried to run a "real" SEV-ES VM and gave up because I needed the "right" OVMF build, blah blah blah. At some point I'll probably bite the bullet and get a "full" CoCo setup working, but it's not exactly at the top of my todo list, in no small part because the triage and debug experience when things go wrong is miles and miles better in selftests.
