On Fri, Jan 05, 2024, Rick P Edgecombe wrote: > On Fri, 2024-01-05 at 10:09 -0800, Jim Mattson wrote: > > > 3. Task switching > > > > Sigh. KVM is forced to emulate task switch, because the hardware is > > incapable of virtualizing it. How hard would it be to make KVM's > > task-switch emulation CET-aware? > > (I am not too familiar with this part of the arch). > > See SDM Vol 3a, chapter 7.3, number 8 and 15. The behavior is around > actual task switching. At first glance, it looks annoying at least. It > would need to do a CMPXCHG to guest memory at some points and take care > to not implement the "Complex Shadow-Stack Updates" behavior. > > But, would anyone use it? I'm not aware of any 32 bit supervisor shadow > stack support out there. So maybe it is ok to just punt to userspace in > this case? Yeah, I think KVM can punt.