Linux Kernel Selftest development
 help / color / mirror / Atom feed
From: Ivan Orlov <iorlov@amazon.com>
To: <bp@alien8.de>, <dave.hansen@linux.intel.com>, <mingo@redhat.com>,
	<pbonzini@redhat.com>, <seanjc@google.com>, <shuah@kernel.org>,
	<tglx@linutronix.de>
Cc: Ivan Orlov <iorlov@amazon.com>, <hpa@zytor.com>,
	<kvm@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
	<linux-kselftest@vger.kernel.org>, <x86@kernel.org>,
	<dwmw@amazon.co.uk>, <pdurrant@amazon.co.uk>,
	<jalliste@amazon.co.uk>
Subject: [PATCH v3 4/7] KVM: VMX: Handle vectoring error in check_emulate_instruction
Date: Tue, 17 Dec 2024 18:14:55 +0000	[thread overview]
Message-ID: <20241217181458.68690-5-iorlov@amazon.com> (raw)
In-Reply-To: <20241217181458.68690-1-iorlov@amazon.com>

Move unhandleable vmexit during vectoring error detection
into check_emulate_instruction. Implement the function which prohibits
the emulation if EMULTYPE_PF is set when vectoring, otherwise such a
situation may occur:

1. CPU executes an instruction and hits a #GP
2. While vectoring the #GP, a shadow #PF occurs
3. On vmexit, KVM re-injects #GP
4. KVM emulates because of the write-protected page
5. KVM "successfully" emulates and also detects the #GP
6. KVM synthesizes a #GP, and since #GP has already been injected,
incorrectly escalates to a #DF.

Fix the comment about EMULTYPE_PF as this flag doesn't necessarily
mean MMIO anymore: it can also be set due to the write protection
violation.

Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Ivan Orlov <iorlov@amazon.com>
---
V1 -> V2:
- Detect the unhandleable vectoring error in vmx_check_emulate_instruction
instead of handling it in the common MMU code (which is specific for
cached MMIO)
V2 -> V3:
- Prohibit any emulation during vectoring if it happens due to an
intercepted #PF.

 arch/x86/include/asm/kvm_host.h |  9 +++++++--
 arch/x86/kvm/vmx/vmx.c          | 23 +++++++++--------------
 2 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index de8fb1ab230c..f3a1d050e1d6 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -2019,8 +2019,8 @@ u64 vcpu_tsc_khz(struct kvm_vcpu *vcpu);
  *			VMware backdoor emulation handles select instructions
  *			and reinjects the #GP for all other cases.
  *
- * EMULTYPE_PF - Set when emulating MMIO by way of an intercepted #PF, in which
- *		 case the CR2/GPA value pass on the stack is valid.
+ * EMULTYPE_PF - Set when an intercepted #PF triggers the emulation, in which case
+ *		 the CR2/GPA value pass on the stack is valid.
  *
  * EMULTYPE_COMPLETE_USER_EXIT - Set when the emulator should update interruptibility
  *				 state and inject single-step #DBs after skipping
@@ -2055,6 +2055,11 @@ u64 vcpu_tsc_khz(struct kvm_vcpu *vcpu);
 #define EMULTYPE_COMPLETE_USER_EXIT (1 << 7)
 #define EMULTYPE_WRITE_PF_TO_SP	    (1 << 8)
 
+static inline bool kvm_can_emulate_event_vectoring(int emul_type)
+{
+	return !(emul_type & EMULTYPE_PF);
+}
+
 int kvm_emulate_instruction(struct kvm_vcpu *vcpu, int emulation_type);
 int kvm_emulate_instruction_from_buffer(struct kvm_vcpu *vcpu,
 					void *insn, int insn_len);
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index acc2f0e0a339..89ddbe1175c7 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -1705,6 +1705,12 @@ int vmx_check_emulate_instruction(struct kvm_vcpu *vcpu, int emul_type,
 		kvm_queue_exception(vcpu, UD_VECTOR);
 		return X86EMUL_PROPAGATE_FAULT;
 	}
+
+	/* Check that emulation is possible during event vectoring */
+	if ((to_vmx(vcpu)->idt_vectoring_info & VECTORING_INFO_VALID_MASK) &&
+	    !kvm_can_emulate_event_vectoring(emul_type))
+		return X86EMUL_UNHANDLEABLE_VECTORING;
+
 	return X86EMUL_CONTINUE;
 }
 
@@ -6543,26 +6549,15 @@ static int __vmx_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath)
 		return 0;
 	}
 
-	/*
-	 * Note:
-	 * Do not try to fix EXIT_REASON_EPT_MISCONFIG if it caused by
-	 * delivery event since it indicates guest is accessing MMIO.
-	 * The vm-exit can be triggered again after return to guest that
-	 * will cause infinite loop.
-	 */
 	if ((vectoring_info & VECTORING_INFO_VALID_MASK) &&
 	    (exit_reason.basic != EXIT_REASON_EXCEPTION_NMI &&
 	     exit_reason.basic != EXIT_REASON_EPT_VIOLATION &&
 	     exit_reason.basic != EXIT_REASON_PML_FULL &&
 	     exit_reason.basic != EXIT_REASON_APIC_ACCESS &&
 	     exit_reason.basic != EXIT_REASON_TASK_SWITCH &&
-	     exit_reason.basic != EXIT_REASON_NOTIFY)) {
-		gpa_t gpa = INVALID_GPA;
-
-		if (exit_reason.basic == EXIT_REASON_EPT_MISCONFIG)
-			gpa = vmcs_read64(GUEST_PHYSICAL_ADDRESS);
-
-		kvm_prepare_event_vectoring_exit(vcpu, gpa);
+	     exit_reason.basic != EXIT_REASON_NOTIFY &&
+	     exit_reason.basic != EXIT_REASON_EPT_MISCONFIG)) {
+		kvm_prepare_event_vectoring_exit(vcpu, INVALID_GPA);
 		return 0;
 	}
 
-- 
2.43.0


  parent reply	other threads:[~2024-12-17 18:15 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-12-17 18:14 [PATCH v3 0/7] Enhance event delivery error handling Ivan Orlov
2024-12-17 18:14 ` [PATCH v3 1/7] KVM: x86: Add function for vectoring error generation Ivan Orlov
2024-12-17 18:14 ` [PATCH v3 2/7] KVM: x86: Add emulation status for unhandleable vectoring Ivan Orlov
2024-12-18 18:23   ` Sean Christopherson
2024-12-17 18:14 ` [PATCH v3 3/7] KVM: x86: Unprotect & retry before unhandleable vectoring check Ivan Orlov
2024-12-17 18:14 ` Ivan Orlov [this message]
2024-12-18 18:39   ` [PATCH v3 4/7] KVM: VMX: Handle vectoring error in check_emulate_instruction Sean Christopherson
2024-12-18 22:00     ` Ivan Orlov
2024-12-17 18:14 ` [PATCH v3 5/7] KVM: SVM: " Ivan Orlov
2024-12-18 18:40   ` Sean Christopherson
2024-12-17 18:14 ` [PATCH v3 6/7] selftests: KVM: extract lidt into helper function Ivan Orlov
2024-12-18 18:41   ` Sean Christopherson
2024-12-17 18:14 ` [PATCH v3 7/7] selftests: KVM: Add test case for MMIO during vectoring Ivan Orlov
2024-12-18 18:44   ` Sean Christopherson
2024-12-18 18:44 ` [PATCH v3 0/7] Enhance event delivery error handling Sean Christopherson
2024-12-18 21:57   ` Ivan Orlov
2024-12-19  2:40 ` Sean Christopherson
2024-12-19 23:19   ` Ivan Orlov

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=20241217181458.68690-5-iorlov@amazon.com \
    --to=iorlov@amazon.com \
    --cc=bp@alien8.de \
    --cc=dave.hansen@linux.intel.com \
    --cc=dwmw@amazon.co.uk \
    --cc=hpa@zytor.com \
    --cc=jalliste@amazon.co.uk \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=pdurrant@amazon.co.uk \
    --cc=seanjc@google.com \
    --cc=shuah@kernel.org \
    --cc=tglx@linutronix.de \
    --cc=x86@kernel.org \
    /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