Messages in this thread | | | From | "Edgecombe, Rick P" <> | Subject | Re: [PATCH] KVM: x86/mmu: x86: Don't overflow lpage_info when checking attributes | Date | Wed, 13 Mar 2024 21:23:02 +0000 |
| |
On Wed, 2024-03-13 at 14:11 -0700, Sean Christopherson wrote: > I wouldn't prioritize speed, I would prioritize overall complexity. > And my gut > reaction is that the overall complexity would go up because we'd need > to make > multiple paths aware that lpage_info could be NULL. There are other > side effects > to making something conditionally valid too, e.g. in the unlikely > scenario where > we mucked up the allocation, KVM would silently fall back to 4KiB > mappings, versus > today KVM would explode (bad for production, but good for > development).
Fair enough, I won't hurry up and try. I'm not sure there would be too many places that would have to handle the out-of-bounds case once everything was suitable wrapped up, though.
| |