On 17/09/21 09:28, Peter Zijlstra wrote:
In theory, I like the idea of burying intel_pt inside x86 (and even in
Intel+VMX code for the most part), but the actual implementation is a
bit gross. Because of the whole "KVM can be a module" thing,
ARGH!! we should really fix that. I've heard other archs have made much
better choices here.
I think that's only ARM, and even then it is only because of limitations
of the hardware which mostly apply only if VHE is not in use.
If anything, it's ARM that should support module build in VHE mode
(Linux would still need to know whether it will be running at EL1 or
EL2, but KVM's functionality is as self-contained as on x86 in the VHE
case).
Paolo
