From: Sean Christopherson <seanjc@google.com>
To: Sean Christopherson <seanjc@google.com>,
Paolo Bonzini <pbonzini@redhat.com>
Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
Xin Li <xin@zytor.com>, Dapeng Mi <dapeng1.mi@linux.intel.com>,
Sandipan Das <sandipan.das@amd.com>
Subject: [PATCH 06/18] KVM: x86: Acquire SRCU in WRMSR fastpath iff instruction needs to be skipped
Date: Tue, 5 Aug 2025 12:05:14 -0700 [thread overview]
Message-ID: <20250805190526.1453366-7-seanjc@google.com> (raw)
In-Reply-To: <20250805190526.1453366-1-seanjc@google.com>
Acquire SRCU in the WRMSR fastpath if and only if an instruction needs to
be skipped, i.e. only if the fastpath succeeds. The reasoning in commit
3f2739bd1e0b ("KVM: x86: Acquire SRCU read lock when handling fastpath MSR
writes") about "avoid having to play whack-a-mole" seems sound, but in
hindsight unconditionally acquiring SRCU does more harm than good.
While acquiring/releasing SRCU isn't slow per se, the things that are
_protected_ by kvm->srcu are generally safe to access only in the "slow"
VM-Exit path. E.g. accessing memslots in generic helpers is never safe,
because accessing guest memory with IRQs disabled is unless unsafe (except
when kvm_vcpu_read_guest_atomic() is used, but that API should never be
used in emulation helpers).
In other words, playing whack-a-mole is actually desirable in this case,
because every access to an asset protected by kvm->srcu warrants further
scrutiny.
Signed-off-by: Sean Christopherson <seanjc@google.com>
---
arch/x86/kvm/x86.c | 21 ++++++++-------------
1 file changed, 8 insertions(+), 13 deletions(-)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 63ca9185d133..69c668f4d2b6 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2158,10 +2158,8 @@ fastpath_t handle_fastpath_set_msr_irqoff(struct kvm_vcpu *vcpu)
{
u32 msr = kvm_rcx_read(vcpu);
u64 data;
- fastpath_t ret;
bool handled;
-
- kvm_vcpu_srcu_read_lock(vcpu);
+ int r;
switch (msr) {
case APIC_BASE_MSR + (APIC_ICR >> 4):
@@ -2177,19 +2175,16 @@ fastpath_t handle_fastpath_set_msr_irqoff(struct kvm_vcpu *vcpu)
break;
}
- if (handled) {
- if (!kvm_skip_emulated_instruction(vcpu))
- ret = EXIT_FASTPATH_EXIT_USERSPACE;
- else
- ret = EXIT_FASTPATH_REENTER_GUEST;
- trace_kvm_msr_write(msr, data);
- } else {
- ret = EXIT_FASTPATH_NONE;
- }
+ if (!handled)
+ return EXIT_FASTPATH_NONE;
+ kvm_vcpu_srcu_read_lock(vcpu);
+ r = kvm_skip_emulated_instruction(vcpu);
kvm_vcpu_srcu_read_unlock(vcpu);
- return ret;
+ trace_kvm_msr_write(msr, data);
+
+ return r ? EXIT_FASTPATH_REENTER_GUEST : EXIT_FASTPATH_EXIT_USERSPACE;
}
EXPORT_SYMBOL_GPL(handle_fastpath_set_msr_irqoff);
--
2.50.1.565.gc32cd1483b-goog
next prev parent reply other threads:[~2025-08-05 19:05 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-08-05 19:05 [PATCH 00/18] KVM: x86: Fastpath cleanups and PMU prep work Sean Christopherson
2025-08-05 19:05 ` [PATCH 01/18] KVM: SVM: Skip fastpath emulation on VM-Exit if next RIP isn't valid Sean Christopherson
2025-08-05 19:05 ` [PATCH 02/18] KVM: x86: Add kvm_icr_to_lapic_irq() helper to allow for fastpath IPIs Sean Christopherson
2025-08-05 19:05 ` [PATCH 03/18] KVM: x86: Only allow "fast" IPIs in fastpath WRMSR(X2APIC_ICR) handler Sean Christopherson
2025-08-05 19:05 ` [PATCH 04/18] KVM: x86: Drop semi-arbitrary restrictions on IPI type in fastpath Sean Christopherson
2025-08-05 19:05 ` [PATCH 05/18] KVM: x86: Unconditionally handle MSR_IA32_TSC_DEADLINE in fastpath exits Sean Christopherson
2025-08-06 17:42 ` Sean Christopherson
2025-08-05 19:05 ` Sean Christopherson [this message]
2025-08-05 19:05 ` [PATCH 07/18] KVM: x86: Unconditionally grab data from EDX:EAX in WRMSR fastpath Sean Christopherson
2025-08-05 19:05 ` [PATCH 08/18] KVM: x86: Fold WRMSR fastpath helpers into the main handler Sean Christopherson
2025-08-05 19:05 ` [PATCH 09/18] KVM: x86/pmu: Move kvm_init_pmu_capability() to pmu.c Sean Christopherson
2025-08-06 7:23 ` Mi, Dapeng
2025-08-05 19:05 ` [PATCH 10/18] KVM: x86/pmu: Add wrappers for counting emulated instructions/branches Sean Christopherson
2025-08-06 7:25 ` Mi, Dapeng
2025-08-05 19:05 ` [PATCH 11/18] KVM: x86/pmu: Calculate set of to-be-emulated PMCs at time of WRMSRs Sean Christopherson
2025-08-06 7:28 ` Mi, Dapeng
2025-08-05 19:05 ` [PATCH 12/18] KVM: x86/pmu: Rename pmc_speculative_in_use() to pmc_is_locally_enabled() Sean Christopherson
2025-08-06 7:28 ` Mi, Dapeng
2025-08-05 19:05 ` [PATCH 13/18] KVM: x86/pmu: Open code pmc_event_is_allowed() in its callers Sean Christopherson
2025-08-06 7:30 ` Mi, Dapeng
2025-08-05 19:05 ` [PATCH 14/18] KVM: x86/pmu: Drop redundant check on PMC being globally enabled for emulation Sean Christopherson
2025-08-06 7:32 ` Mi, Dapeng
2025-08-05 19:05 ` [PATCH 15/18] KVM: x86/pmu: Drop redundant check on PMC being locally " Sean Christopherson
2025-08-06 7:33 ` Mi, Dapeng
2025-08-05 19:05 ` [PATCH 16/18] KVM: x86/pmu: Rename check_pmu_event_filter() to pmc_is_event_allowed() Sean Christopherson
2025-08-06 7:35 ` Mi, Dapeng
2025-08-05 19:05 ` [PATCH 17/18] KVM: x86: Push acquisition of SRCU in fastpath into kvm_pmu_trigger_event() Sean Christopherson
2025-08-06 8:08 ` Mi, Dapeng
2025-08-06 17:33 ` Sean Christopherson
2025-08-07 2:24 ` Mi, Dapeng
2025-08-07 13:31 ` Sean Christopherson
2025-08-08 0:42 ` Mi, Dapeng
2025-08-05 19:05 ` [PATCH 18/18] KVM: x86: Add a fastpath handler for INVD Sean Christopherson
2025-08-06 8:11 ` [PATCH 00/18] KVM: x86: Fastpath cleanups and PMU prep work Mi, Dapeng
2025-08-07 6:23 ` Sandipan Das
2025-08-19 23:11 ` Sean Christopherson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20250805190526.1453366-7-seanjc@google.com \
--to=seanjc@google.com \
--cc=dapeng1.mi@linux.intel.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=sandipan.das@amd.com \
--cc=xin@zytor.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®