Messages in this thread | | | Date | Tue, 11 Jan 2022 16:15:36 +0800 | From | Like Xu <> | Subject | Re: [PATCH v2] KVM: x86/pt: Ignore all unknown Intel PT capabilities |
| |
On 11/1/2022 3:59 pm, Xiaoyao Li wrote: >> >> +#define GUEST_SUPPORTED_CPUID_14_EBX \ >> + (BIT(0) | BIT(1) | BIT(2) | BIT(3) | BIT(4) | BIT(5)) >> + >> +#define GUEST_SUPPORTED_CPUID_14_ECX \ >> + (BIT(0) | BIT(1) | BIT(2) | BIT(3) | BIT(31)) >> + > > I doubt BIT(3) of CPUID_14_ECX can be exposed to guest directly. > > It means "output to Trace Transport Subsystem Supported". If I understand > correctly, it at least needs passthrough of the said Transport Subsystem or > emulation of it.
I'm not surprised that we can route Intel Guest PT output to a platform-specific trace endpoint (e.g., physical or emulated JTAG) as an MMIO debug port.
| |