On 04/10/2018 12:20, Alexander Shishkin wrote: > Paolo Bonzini <pbonzini@xxxxxxxxxx> writes: > >> On 02/10/2018 12:17, Alexander Shishkin wrote: >>> Ok, so this is where the host PT is disabled. >>> >>> So, my recommendation for this patchset is: we need to address the fact >>> that we are disabling host PT without telling the host. We need to >>> either stop doing it or inform the host. I don't think there need to be >>> any more versions of this patchset until this issue is addressed. >> >> If you don't want that effect, do not enable host-guest mode. The >> default is "off" in fact. > > One shouldn't have to enable or disable anything in KVM to stop it from > breaking one's existing workflow. That makes no sense. If you "have to enable or disable anything" it means you have to override the default. But the default in this patches is "no change compared to before the patches", leaving tracing of both host and guest entirely to the host, so I don't understand your remark. What workflow is broken? > There already are controls in perf that enable/disable guest tracing. You are confusing "tracing guest from the host" and "the guest can trace itself". This patchset is adding support for the latter, and that affects directly whether the tracing CPUID leaf can be added to the guest. Therefore it's not perf that can decide whether to turn it on; KVM must know it when /dev/kvm is opened, which is why it is a module parameter. Paolo
