[PATCH] perf/x86/amd: Don't touch the Host-Only/Guest-Only bit inside the guest

From: Dongli Si
Date: Fri Apr 01 2022 - 04:37:51 EST


From: Dongli Si <sidongli1997@xxxxxxxxx>

With nested virtualization on AMD Milan:

If "perf record" is run in an L1 hypervisor with an L2 guest,
the following warning is emitted in the L1 guest.

[] unchecked MSR access error: WRMSR to 0xc0010200 (tried to write 0x0000020000510076)
at rIP: 0xffffffff81003a50 (x86_pmu_enable_all+0x60/0x100)
[] Call Trace:
[] <IRQ>
[] ? x86_pmu_enable+0x146/0x300
[] __perf_install_in_context+0x150/0x170

and

If "perf record" is run in an L2 guest,
the following warning is emitted in the L1 guest.

[] unchecked MSR access error: WRMSR to 0xc0010200 (tried to write 0x0000010000510076)
at rIP: 0xffffffff81003a50 (x86_pmu_enable_all+0x60/0x100)
[] Call Trace:
[] <TASK>
[] ? x86_pmu_enable+0x146/0x300
[] __perf_install_in_context+0x150/0x170

The AMD64_EVENTSEL_HOSTONLY/AMD64_EVENTSEL_GUESTONLY bit is defined and
used on the host (L0), while the L1 hypervisor Performance Monitor Unit
should avoid such use.

Fixes: 1018faa6cf23 ("perf/x86/kvm: Fix Host-Only/Guest-Only counting with SVM disabled")
Signed-off-by: Dongli Si <sidongli1997@xxxxxxxxx>
---
arch/x86/events/amd/core.c | 23 ++++++++++++++++++++++-
1 file changed, 22 insertions(+), 1 deletion(-)

diff --git a/arch/x86/events/amd/core.c b/arch/x86/events/amd/core.c
index 9687a8aef01c..79bd60d6b0fd 100644
--- a/arch/x86/events/amd/core.c
+++ b/arch/x86/events/amd/core.c
@@ -8,6 +8,7 @@
#include <linux/jiffies.h>
#include <asm/apicdef.h>
#include <asm/nmi.h>
+#include <asm/hypervisor.h>

#include "../perf_event.h"

@@ -533,7 +534,15 @@ static void amd_pmu_cpu_starting(int cpu)
struct amd_nb *nb;
int i, nb_id;

- cpuc->perf_ctr_virt_mask = AMD64_EVENTSEL_HOSTONLY;
+ /*
+ * The Host-Only/Guest-Only bit can only be set on the host.
+ * When SVM is disabled, set the Host-Only bit will cause the
+ * performance counter to not work.
+ */
+ if (hypervisor_is_type(X86_HYPER_NATIVE))
+ cpuc->perf_ctr_virt_mask = AMD64_EVENTSEL_HOSTONLY;
+ else
+ cpuc->perf_ctr_virt_mask = AMD64_EVENTSEL_GUESTONLY | AMD64_EVENTSEL_HOSTONLY;

if (!x86_pmu.amd_nb_constraints)
return;
@@ -1023,10 +1032,16 @@ __init int amd_pmu_init(void)
return 0;
}

+/*
+ * Unmask the Host-only bit when virtualization is enabled on the Host Hypervisor
+ */
void amd_pmu_enable_virt(void)
{
struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);

+ if (!hypervisor_is_type(X86_HYPER_NATIVE))
+ return;
+
cpuc->perf_ctr_virt_mask = 0;

/* Reload all events */
@@ -1035,10 +1050,16 @@ void amd_pmu_enable_virt(void)
}
EXPORT_SYMBOL_GPL(amd_pmu_enable_virt);

+/*
+ * Mask the Host-only bit when virtualization is disabled on the Host Hypervisor
+ */
void amd_pmu_disable_virt(void)
{
struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);

+ if (!hypervisor_is_type(X86_HYPER_NATIVE))
+ return;
+
/*
* We only mask out the Host-only bit so that host-only counting works
* when SVM is disabled. If someone sets up a guest-only counter when
--
2.32.0