Skip to content

Commit

Permalink
KVM: vmx/pmu: Expose LBR_FMT in the MSR_IA32_PERF_CAPABILITIES
Browse files Browse the repository at this point in the history
Userspace could enable guest LBR feature when the exactly supported
LBR format value is initialized to the MSR_IA32_PERF_CAPABILITIES
and the LBR is also compatible with vPMU version and host cpu model.

The LBR could be enabled on the guest if host perf supports LBR
(checked via x86_perf_get_lbr()) and the vcpu model is compatible
with the host one.

Signed-off-by: Like Xu <like.xu@linux.intel.com>
Message-Id: <20210201051039.255478-11-like.xu@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
  • Loading branch information
Like Xu authored and Paolo Bonzini committed Feb 4, 2021
1 parent 9aa4f62 commit be635e3
Showing 1 changed file with 8 additions and 1 deletion.
9 changes: 8 additions & 1 deletion arch/x86/kvm/vmx/capabilities.h
Original file line number Diff line number Diff line change
Expand Up @@ -380,11 +380,18 @@ static inline bool vmx_pt_mode_is_host_guest(void)

static inline u64 vmx_get_perf_capabilities(void)
{
u64 perf_cap = 0;

if (boot_cpu_has(X86_FEATURE_PDCM))
rdmsrl(MSR_IA32_PERF_CAPABILITIES, perf_cap);

perf_cap &= PMU_CAP_LBR_FMT;

/*
* Since counters are virtualized, KVM would support full
* width counting unconditionally, even if the host lacks it.
*/
return PMU_CAP_FW_WRITES;
return PMU_CAP_FW_WRITES | perf_cap;
}

static inline u64 vmx_supported_debugctl(void)
Expand Down

0 comments on commit be635e3

Please sign in to comment.