linux-kernel - [PATCH] KVM: vPMU: Don't program counter for interrupt-based event sampling w/o lapic_in

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <1634894233-84041-1-git-send-email-wanpengli@tencent.com>
Date:   Fri, 22 Oct 2021 02:17:13 -0700
From:   Wanpeng Li <kernellwp@...il.com>
To:     linux-kernel@...r.kernel.org, kvm@...r.kernel.org
Cc:     Paolo Bonzini <pbonzini@...hat.com>,
        Sean Christopherson <seanjc@...gle.com>,
        Vitaly Kuznetsov <vkuznets@...hat.com>,
        Wanpeng Li <wanpengli@...cent.com>,
        Jim Mattson <jmattson@...gle.com>,
        Joerg Roedel <joro@...tes.org>
Subject: [PATCH] KVM: vPMU: Don't program counter for interrupt-based event sampling w/o lapic_in_kernel

From: Wanpeng Li <wanpengli@...cent.com>

vPMU depends on in-kernel lapic to deliver pmi interrupt, there is a
lot of overhead when creating/maintaining perf_event object, 
locking/unlocking perf_event_ctx etc for vPMU. It silently fails to 
deliver pmi interrupt if w/o in-kernel lapic currently. Let's not 
program counter for interrupt-based event sampling w/o in-kernel 
lapic support to avoid the whole bothering. 

Signed-off-by: Wanpeng Li <wanpengli@...cent.com>
---
 arch/x86/kvm/pmu.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kvm/pmu.c b/arch/x86/kvm/pmu.c
index 0772bad9165c..fa5cd33af10d 100644
--- a/arch/x86/kvm/pmu.c
+++ b/arch/x86/kvm/pmu.c
@@ -179,6 +179,7 @@ void reprogram_gp_counter(struct kvm_pmc *pmc, u64 eventsel)
 	struct kvm_pmu_event_filter *filter;
 	int i;
 	bool allow_event = true;
+	bool intr = eventsel & ARCH_PERFMON_EVENTSEL_INT;
 
 	if (eventsel & ARCH_PERFMON_EVENTSEL_PIN_CONTROL)
 		printk_once("kvm pmu: pin control bit is ignored\n");
@@ -187,7 +188,8 @@ void reprogram_gp_counter(struct kvm_pmc *pmc, u64 eventsel)
 
 	pmc_pause_counter(pmc);
 
-	if (!(eventsel & ARCH_PERFMON_EVENTSEL_ENABLE) || !pmc_is_enabled(pmc))
+	if (!(eventsel & ARCH_PERFMON_EVENTSEL_ENABLE) || !pmc_is_enabled(pmc)
+	    || (intr && !lapic_in_kernel(pmc->vcpu)))
 		return;
 
 	filter = srcu_dereference(kvm->arch.pmu_event_filter, &kvm->srcu);
@@ -233,7 +235,7 @@ void reprogram_gp_counter(struct kvm_pmc *pmc, u64 eventsel)
 	pmc_reprogram_counter(pmc, type, config,
 			      !(eventsel & ARCH_PERFMON_EVENTSEL_USR),
 			      !(eventsel & ARCH_PERFMON_EVENTSEL_OS),
-			      eventsel & ARCH_PERFMON_EVENTSEL_INT,
+			      intr,
 			      (eventsel & HSW_IN_TX),
 			      (eventsel & HSW_IN_TX_CHECKPOINTED));
 }
@@ -248,7 +250,7 @@ void reprogram_fixed_counter(struct kvm_pmc *pmc, u8 ctrl, int idx)
 
 	pmc_pause_counter(pmc);
 
-	if (!en_field || !pmc_is_enabled(pmc))
+	if (!en_field || !pmc_is_enabled(pmc) || (pmi && !lapic_in_kernel(pmc->vcpu)))
 		return;
 
 	filter = srcu_dereference(kvm->arch.pmu_event_filter, &kvm->srcu);
-- 
2.25.1