Re: [PATCH v10 16/29] KVM: selftests: Test Intel PMU architectural events on gp counters

From: Mi, Dapeng
Date: Sun Jan 14 2024 - 21:03:25 EST



On 1/13/2024 5:37 AM, Sean Christopherson wrote:
On Fri, Jan 12, 2024, Dapeng Mi wrote:
On 1/10/2024 7:02 AM, Sean Christopherson wrote:
+/*
+ * If an architectural event is supported and guaranteed to generate at least
+ * one "hit, assert that its count is non-zero. If an event isn't supported or
+ * the test can't guarantee the associated action will occur, then all bets are
+ * off regarding the count, i.e. no checks can be done.
+ *
+ * Sanity check that in all cases, the event doesn't count when it's disabled,
+ * and that KVM correctly emulates the write of an arbitrary value.
+ */
+static void guest_assert_event_count(uint8_t idx,
+ struct kvm_x86_pmu_feature event,
+ uint32_t pmc, uint32_t pmc_msr)
+{
+ uint64_t count;
+
+ count = _rdpmc(pmc);
+ if (!this_pmu_has(event))
+ goto sanity_checks;
+
+ switch (idx) {
+ case INTEL_ARCH_INSTRUCTIONS_RETIRED_INDEX:
+ GUEST_ASSERT_EQ(count, NUM_INSNS_RETIRED);
+ break;
+ case INTEL_ARCH_BRANCHES_RETIRED_INDEX:
+ GUEST_ASSERT_EQ(count, NUM_BRANCHES);
+ break;
+ case INTEL_ARCH_CPU_CYCLES_INDEX:
+ case INTEL_ARCH_REFERENCE_CYCLES_INDEX:
Since we already support slots event in below guest_test_arch_event(), we
can add check for INTEL_ARCH_TOPDOWN_SLOTS_INDEX here.
Can that actually be tested at this point, since KVM doesn't support
X86_PMU_FEATURE_TOPDOWN_SLOTS, i.e. this_pmu_has() above should always fail, no?

I suppose X86_PMU_FEATURE_TOPDOWN_SLOTS has been supported in KVM.  The following output comes from a guest with latest kvm-x86 code on the Sapphire Rapids platform.

sudo cpuid -l 0xa
CPU 0:
   Architecture Performance Monitoring Features (0xa):
      version ID                               = 0x2 (2)
      number of counters per logical processor = 0x8 (8)
      bit width of counter                     = 0x30 (48)
      length of EBX bit vector                 = 0x8 (8)
      core cycle event                         = available
      instruction retired event                = available
      reference cycles event                   = available
      last-level cache ref event               = available
      last-level cache miss event              = available
      branch inst retired event                = available
      branch mispred retired event             = available
      top-down slots event                     = available

Current KVM doesn't support fixed counter 3 and pseudo slots event yet, but the architectural slots event is supported and can be programed on a GP counter. Current test code can cover this case, so I think we'd better add the check for the slots count.



I'm hesitant to add an assertion of any king without the ability to actually test
the code.