On Wed, 2023-09-13 at 16:41 +0800, Like Xu wrote: > On 13/9/2023 4:10 pm, David Woodhouse wrote: > > On Fri, 2023-08-11 at 15:59 -0700, Sean Christopherson wrote: > > > The problem isn't that the sync code doesn't differentiate between kernel and > > > user-initiated writes, because parts of the code *do* differentiate. I think it's > > > more accurate to say that the problem is that the sync code doesn't differentiate > > > between userspace initializing the TSC and userspace attempting to synchronize the > > > TSC. > > > > I'm not utterly sure that *I* differentiate between userspace > > "initializing the TSC" and attempting to "synchronize the TSC". What > > *is* the difference? > > I'd be more inclined to Oliver's explanation in this version of the changelog > that different tsc_offsets are used to calculate guest_tsc value between the vcpu > is created and when it is first set by usersapce. This extra synchronization is not > expected for guest based on user's bugzilla report. > Yes, it's about the kernel's default startup values (first vCPU starting at TSC 0, others syncing to that on creation), and the fact that the *first* userspace write (to any vCPU) should actually be honoured even if it *does* happen to be within 1 second of the kernel's startup values. > Two hands in favor. Using the new KVM_VCPU_TSC_OFFSET API and a little > fix on the legacy API is not conflict. Thank you for reviewing it. I'm slightly dubious about making *changes* to an established userspace ABI, especially when there's already a better way to do it. But I suppose this specific change, if you *don't* also take away the ability for userspace to explicitly write zero to force a sync (qv), is OK.
Attachment:
smime.p7s
Description: S/MIME cryptographic signature
