Re: [PATCH] KVM: arm64: Disable TRBE Trace Buffer Unit when running in guest context

On Mon, Feb 16, 2026 at 04:49:02PM +0000, Marc Zyngier wrote:

[...]

quoted

quoted

But that's *not* working. If you trace EL1 only, even with a VHE host,
the result is not usable.

Do you mean not working because of the missing exceptions? I did a bit
of testing before and the trace did seem somewhat usable to me. It had
EL1 and EL0 atoms in there.

Sure. Now try to look at what that means for NV, where all the
EL1->EL2 exceptions are emulated, where all the EL2->EL1 exception
returns are emulated.

What does it give you? A bag of nonsense.

Sorry for jumping in.

If we enable TRBE in a VM, whether nested or not, why is it necessary to
capture trace data for the exception transition between the VM and its
higher level host(s)?

Seems to me, regardless of what happens during exception emulation, once
the VM is switched out, tracing will be stopped, and then re-enabled
when the VM is switched back.  In that case, we should be able to record
the complete trace data for whatever occurs while the VM is running.

On the other hand, when launch a trace within a VM, I think we should
not trace higher level's hypervisor or hosts, this is concerned for
security leakage.

Same thing for EL2->EL0, by the way, so you can't even correctly
profile an EL0 program that performs a syscall, or that gets
interrupted. And while without NV, these exceptions are rare, having a
trace that is unreliable has the potential of being worse than no
trace at all.

Until the architecture grows a way for KVM to inject the missing
information into the trace, TRBE support for guest will stay out.

I agree we need to understand what is actually blocking issues for TRBE
virtualization.

Essentially, I'd confirm the methodology for trace virtualization.  I
assume it allows a higher privilege OS to trace a lower privilege OS,
but not the other way around.

Thanks,
Leo