Skip to contentSkip to conversation

perf/x86/core: Completely disable guest PEBS via guest's global_ctrl

PATCH RESEND Applied Like Xu 3 participants 3 messages Aug 31, 2022 – Sep 2, 2022

Subsystems (3)
lore
Share & tools
lore

Like Xu

author
#0 lore raw

PATCH RESEND 1/1 perf/x86/core: Completely disable guest PEBS via guest's global_ctrl

From: Like Xu <[email protected]>

When a guest PEBS counter is cross-mapped by a host counter, software will remove the corresponding bit in the arr[global_ctrl].guest and expect hardware to perform a change of state "from enable to disable" via the msr_slot[] switch during the vmx transaction.

The real world is that if user adjust the counter overflow value small enough, it still opens a tiny race window for the previously PEBS-enabled counter to write cross-mapped PEBS records into the guest's PEBS buffer, when arr[global_ctrl].guest has been prioritised (switch_msr_special stuff) to switch into the enabled state, while the arr[pebs_enable].guest has not.

Close this window by clearing invalid bits in the arr[global_ctrl].guest.

Cc Kan Liang <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Sean Christopherson <[email protected]>
Fixes 854250329c02 ("KVM: x86/pmu: Disable guest PEBS temporarily in two rare situations")
---
 arch/x86/events/intel/core.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
arch/x86/events/intel/core.c +2 -1
index 2db93498ff71..75cdd11ab014 100644
@@ -4052,8 +4052,9 @@ static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr, void *data)
 		/* Disable guest PEBS if host PEBS is enabled. */
 		arr[pebs_enable].guest = 0;
 	} else {
-		/* Disable guest PEBS for cross-mapped PEBS counters. */
+		/* Disable guest PEBS thoroughly for cross-mapped PEBS counters. */
 		arr[pebs_enable].guest &= ~kvm_pmu->host_cross_mapped_mask;
+		arr[global_ctrl].guest &= ~kvm_pmu->host_cross_mapped_mask;
 		/* Set hw GLOBAL_CTRL bits for PEBS counter when it runs for guest */
 		arr[global_ctrl].guest |= arr[pebs_enable].guest;
 	}
Signature · 1 line
2.37.3
maintainer
#1 lore raw
Quoted 15 lines · Like Xu

From: Like Xu <[email protected]>

When a guest PEBS counter is cross-mapped by a host counter, software will remove the corresponding bit in the arr[global_ctrl].guest and expect hardware to perform a change of state "from enable to disable" via the msr_slot[] switch during the vmx transaction.

The real world is that if user adjust the counter overflow value small enough, it still opens a tiny race window for the previously PEBS-enabled counter to write cross-mapped PEBS records into the guest's PEBS buffer, when arr[global_ctrl].guest has been prioritised (switch_msr_special stuff) to switch into the enabled state, while the arr[pebs_enable].guest has not.

Close this window by clearing invalid bits in the arr[global_ctrl].guest.

Quoted 7 lines
Cc Kan Liang <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Sean Christopherson <[email protected]>
Fixes 854250329c02 ("KVM: x86/pmu: Disable guest PEBS temporarily in two rare situations")

---

Thanks!

merge bot Applied
#2 lore raw
In reply to Like Xu
The following commit has been merged into the perf/urgent branch of tip:

The following commit has been merged into the perf/urgent branch of tip:

Commit-ID:     f2aeea57504cbbc58da3c59b939fc16150087648
Gitweb:        https://git.kernel.org/tip/f2aeea57504cbbc58da3c59b939fc16150087648
Author:        Like Xu <[email protected]>
AuthorDate:    Wed, 31 Aug 2022 11:35:24 +08:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Thu, 01 Sep 2022 11:19:42 +02:00

perf/x86/core: Completely disable guest PEBS via guest's global_ctrl

When a guest PEBS counter is cross-mapped by a host counter, software will remove the corresponding bit in the arr[global_ctrl].guest and expect hardware to perform a change of state "from enable to disable" via the msr_slot[] switch during the vmx transaction.

The real world is that if user adjust the counter overflow value small enough, it still opens a tiny race window for the previously PEBS-enabled counter to write cross-mapped PEBS records into the guest's PEBS buffer, when arr[global_ctrl].guest has been prioritised (switch_msr_special stuff) to switch into the enabled state, while the arr[pebs_enable].guest has not.

Close this window by clearing invalid bits in the arr[global_ctrl].guest.

Fixes 854250329c02 ("KVM: x86/pmu: Disable guest PEBS temporarily in two rare situations")
---
 arch/x86/events/intel/core.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
arch/x86/events/intel/core.c +2 -1
index e8eb936..c20d8cd 100644
@@ -4061,8 +4061,9 @@ static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr, void *data)
 		/* Disable guest PEBS if host PEBS is enabled. */
 		arr[pebs_enable].guest = 0;
 	} else {
-		/* Disable guest PEBS for cross-mapped PEBS counters. */
+		/* Disable guest PEBS thoroughly for cross-mapped PEBS counters. */
 		arr[pebs_enable].guest &= ~kvm_pmu->host_cross_mapped_mask;
+		arr[global_ctrl].guest &= ~kvm_pmu->host_cross_mapped_mask;
 		/* Set hw GLOBAL_CTRL bits for PEBS counter when it runs for guest */
 		arr[global_ctrl].guest |= arr[pebs_enable].guest;
 	}