Received-SPF: pass (google.com: domain of linux-kernel+bounces-81648-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
From: Lai Jiangshan <jiangshanlai@gmail.com>
To: linux-kernel@vger.kernel.org
Cc: Lai Jiangshan <jiangshan.ljs@antgroup.com>,
	Hou Wenlong <houwenlong.hwl@antgroup.com>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Sean Christopherson <seanjc@google.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Borislav Petkov <bp@alien8.de>,
	Ingo Molnar <mingo@redhat.com>,
	kvm@vger.kernel.org,
	Paolo Bonzini <pbonzini@redhat.com>,
	x86@kernel.org,
	Kees Cook <keescook@chromium.org>,
	Juergen Gross <jgross@suse.com>,
	Dave Hansen <dave.hansen@linux.intel.com>,
	"H. Peter Anvin" <hpa@zytor.com>,
	"Mike Rapoport (IBM)" <rppt@kernel.org>,
	Yu-cheng Yu <yu-cheng.yu@intel.com>,
	Rick Edgecombe <rick.p.edgecombe@intel.com>,
	"Paul E. McKenney" <paulmck@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>
Subject: [RFC PATCH 06/73] KVM: x86: Move VMX interrupt/nmi handling into kvm.ko
Date: Mon, 26 Feb 2024 22:35:23 +0800
Message-Id: <20240226143630.33643-7-jiangshanlai@gmail.com>
In-Reply-To: <20240226143630.33643-1-jiangshanlai@gmail.com>
References: <20240226143630.33643-1-jiangshanlai@gmail.com>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

From: Lai Jiangshan <jiangshan.ljs@antgroup.com>

Similar to VMX, hardware interrupts/NMI during guest running in PVM will
trigger VM exit and should be handled by host interrupt/NMI handlers.
Therefore, move VMX interrupt/NMI handling into kvm.ko for common usage.

Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
Co-developed-by: Hou Wenlong <houwenlong.hwl@antgroup.com>
Signed-off-by: Hou Wenlong <houwenlong.hwl@antgroup.com>
---
 arch/x86/include/asm/idtentry.h | 12 ++++----
 arch/x86/kernel/nmi.c           |  8 +++---
 arch/x86/kvm/Makefile           |  2 +-
 arch/x86/kvm/host_entry.S       | 50 +++++++++++++++++++++++++++++++++
 arch/x86/kvm/vmx/vmenter.S      | 43 ----------------------------
 arch/x86/kvm/vmx/vmx.c          | 14 ++-------
 arch/x86/kvm/x86.c              |  3 ++
 arch/x86/kvm/x86.h              | 18 ++++++++++++
 8 files changed, 85 insertions(+), 65 deletions(-)
 create mode 100644 arch/x86/kvm/host_entry.S

diff --git a/arch/x86/include/asm/idtentry.h b/arch/x86/include/asm/idtentry.h
index 13639e57e1f8..8aab0b50431a 100644
--- a/arch/x86/include/asm/idtentry.h
+++ b/arch/x86/include/asm/idtentry.h
@@ -586,14 +586,14 @@ DECLARE_IDTENTRY_RAW(X86_TRAP_MC,	xenpv_exc_machine_check);
 
 /* NMI */
 
-#if IS_ENABLED(CONFIG_KVM_INTEL)
+#if IS_ENABLED(CONFIG_KVM)
 /*
- * Special entry point for VMX which invokes this on the kernel stack, even for
- * 64-bit, i.e. without using an IST.  asm_exc_nmi() requires an IST to work
- * correctly vs. the NMI 'executing' marker.  Used for 32-bit kernels as well
- * to avoid more ifdeffery.
+ * Special entry point for VMX/PVM which invokes this on the kernel stack, even
+ * for 64-bit, i.e. without using an IST.  asm_exc_nmi() requires an IST to
+ * work correctly vs. the NMI 'executing' marker.  Used for 32-bit kernels as
+ * well to avoid more ifdeffery.
  */
-DECLARE_IDTENTRY(X86_TRAP_NMI,		exc_nmi_kvm_vmx);
+DECLARE_IDTENTRY(X86_TRAP_NMI,		exc_nmi_kvm);
 #endif
 
 DECLARE_IDTENTRY_NMI(X86_TRAP_NMI,	exc_nmi);
diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
index 17e955ab69fe..265e6b38cc58 100644
--- a/arch/x86/kernel/nmi.c
+++ b/arch/x86/kernel/nmi.c
@@ -568,13 +568,13 @@ DEFINE_IDTENTRY_RAW(exc_nmi)
 		mds_user_clear_cpu_buffers();
 }
 
-#if IS_ENABLED(CONFIG_KVM_INTEL)
-DEFINE_IDTENTRY_RAW(exc_nmi_kvm_vmx)
+#if IS_ENABLED(CONFIG_KVM)
+DEFINE_IDTENTRY_RAW(exc_nmi_kvm)
 {
 	exc_nmi(regs);
 }
-#if IS_MODULE(CONFIG_KVM_INTEL)
-EXPORT_SYMBOL_GPL(asm_exc_nmi_kvm_vmx);
+#if IS_MODULE(CONFIG_KVM)
+EXPORT_SYMBOL_GPL(asm_exc_nmi_kvm);
 #endif
 #endif
 
diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile
index 80e3fe184d17..97bad203b1b1 100644
--- a/arch/x86/kvm/Makefile
+++ b/arch/x86/kvm/Makefile
@@ -9,7 +9,7 @@ endif
 
 include $(srctree)/virt/kvm/Makefile.kvm
 
-kvm-y			+= x86.o emulate.o i8259.o irq.o lapic.o \
+kvm-y			+= x86.o emulate.o i8259.o irq.o lapic.o host_entry.o \
 			   i8254.o ioapic.o irq_comm.o cpuid.o pmu.o mtrr.o \
 			   hyperv.o debugfs.o mmu/mmu.o mmu/page_track.o \
 			   mmu/spte.o
diff --git a/arch/x86/kvm/host_entry.S b/arch/x86/kvm/host_entry.S
new file mode 100644
index 000000000000..6bdf0df06eb0
--- /dev/null
+++ b/arch/x86/kvm/host_entry.S
@@ -0,0 +1,50 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#include <linux/linkage.h>
+#include <asm/asm.h>
+#include <asm/nospec-branch.h>
+#include <asm/segment.h>
+
+.macro KVM_DO_EVENT_IRQOFF call_insn call_target
+	/*
+	 * Unconditionally create a stack frame, getting the correct RSP on the
+	 * stack (for x86-64) would take two instructions anyways, and RBP can
+	 * be used to restore RSP to make objtool happy (see below).
+	 */
+	push %_ASM_BP
+	mov %_ASM_SP, %_ASM_BP
+
+#ifdef CONFIG_X86_64
+	/*
+	 * Align RSP to a 16-byte boundary (to emulate CPU behavior) before
+	 * creating the synthetic interrupt stack frame for the IRQ/NMI.
+	 */
+	and  $-16, %rsp
+	push $__KERNEL_DS
+	push %rbp
+#endif
+	pushf
+	push $__KERNEL_CS
+	\call_insn \call_target
+
+	/*
+	 * "Restore" RSP from RBP, even though IRET has already unwound RSP to
+	 * the correct value.  objtool doesn't know the callee will IRET and,
+	 * without the explicit restore, thinks the stack is getting walloped.
+	 * Using an unwind hint is problematic due to x86-64's dynamic alignment.
+	 */
+	mov %_ASM_BP, %_ASM_SP
+	pop %_ASM_BP
+	RET
+.endm
+
+.section .noinstr.text, "ax"
+
+SYM_FUNC_START(kvm_do_host_nmi_irqoff)
+	KVM_DO_EVENT_IRQOFF call asm_exc_nmi_kvm
+SYM_FUNC_END(kvm_do_host_nmi_irqoff)
+
+.section .text, "ax"
+
+SYM_FUNC_START(kvm_do_host_interrupt_irqoff)
+	KVM_DO_EVENT_IRQOFF CALL_NOSPEC _ASM_ARG1
+SYM_FUNC_END(kvm_do_host_interrupt_irqoff)
diff --git a/arch/x86/kvm/vmx/vmenter.S b/arch/x86/kvm/vmx/vmenter.S
index 906ecd001511..12b7b99a9dd8 100644
--- a/arch/x86/kvm/vmx/vmenter.S
+++ b/arch/x86/kvm/vmx/vmenter.S
@@ -31,39 +31,6 @@
 #define VCPU_R15	__VCPU_REGS_R15 * WORD_SIZE
 #endif
 
-.macro VMX_DO_EVENT_IRQOFF call_insn call_target
-	/*
-	 * Unconditionally create a stack frame, getting the correct RSP on the
-	 * stack (for x86-64) would take two instructions anyways, and RBP can
-	 * be used to restore RSP to make objtool happy (see below).
-	 */
-	push %_ASM_BP
-	mov %_ASM_SP, %_ASM_BP
-
-#ifdef CONFIG_X86_64
-	/*
-	 * Align RSP to a 16-byte boundary (to emulate CPU behavior) before
-	 * creating the synthetic interrupt stack frame for the IRQ/NMI.
-	 */
-	and  $-16, %rsp
-	push $__KERNEL_DS
-	push %rbp
-#endif
-	pushf
-	push $__KERNEL_CS
-	\call_insn \call_target
-
-	/*
-	 * "Restore" RSP from RBP, even though IRET has already unwound RSP to
-	 * the correct value.  objtool doesn't know the callee will IRET and,
-	 * without the explicit restore, thinks the stack is getting walloped.
-	 * Using an unwind hint is problematic due to x86-64's dynamic alignment.
-	 */
-	mov %_ASM_BP, %_ASM_SP
-	pop %_ASM_BP
-	RET
-.endm
-
 .section .noinstr.text, "ax"
 
 /**
@@ -299,10 +266,6 @@ SYM_INNER_LABEL_ALIGN(vmx_vmexit, SYM_L_GLOBAL)
 
 SYM_FUNC_END(__vmx_vcpu_run)
 
-SYM_FUNC_START(vmx_do_nmi_irqoff)
-	VMX_DO_EVENT_IRQOFF call asm_exc_nmi_kvm_vmx
-SYM_FUNC_END(vmx_do_nmi_irqoff)
-
 #ifndef CONFIG_CC_HAS_ASM_GOTO_OUTPUT
 
 /**
@@ -354,9 +317,3 @@ SYM_FUNC_START(vmread_error_trampoline)
 	RET
 SYM_FUNC_END(vmread_error_trampoline)
 #endif
-
-.section .text, "ax"
-
-SYM_FUNC_START(vmx_do_interrupt_irqoff)
-	VMX_DO_EVENT_IRQOFF CALL_NOSPEC _ASM_ARG1
-SYM_FUNC_END(vmx_do_interrupt_irqoff)
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index be20a60047b1..fca47304506e 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -6920,9 +6920,6 @@ static void vmx_apicv_pre_state_restore(struct kvm_vcpu *vcpu)
 	memset(vmx->pi_desc.pir, 0, sizeof(vmx->pi_desc.pir));
 }
 
-void vmx_do_interrupt_irqoff(unsigned long entry);
-void vmx_do_nmi_irqoff(void);
-
 static void handle_nm_fault_irqoff(struct kvm_vcpu *vcpu)
 {
 	/*
@@ -6968,9 +6965,7 @@ static void handle_external_interrupt_irqoff(struct kvm_vcpu *vcpu)
 	    "unexpected VM-Exit interrupt info: 0x%x", intr_info))
 		return;
 
-	kvm_before_interrupt(vcpu, KVM_HANDLING_IRQ);
-	vmx_do_interrupt_irqoff(gate_offset(desc));
-	kvm_after_interrupt(vcpu);
+	kvm_do_interrupt_irqoff(vcpu, gate_offset(desc));
 
 	vcpu->arch.at_instruction_boundary = true;
 }
@@ -7260,11 +7255,8 @@ static noinstr void vmx_vcpu_enter_exit(struct kvm_vcpu *vcpu,
 		vmx->idt_vectoring_info = vmcs_read32(IDT_VECTORING_INFO_FIELD);
 
 	if ((u16)vmx->exit_reason.basic == EXIT_REASON_EXCEPTION_NMI &&
-	    is_nmi(vmx_get_intr_info(vcpu))) {
-		kvm_before_interrupt(vcpu, KVM_HANDLING_NMI);
-		vmx_do_nmi_irqoff();
-		kvm_after_interrupt(vcpu);
-	}
+	    is_nmi(vmx_get_intr_info(vcpu)))
+		kvm_do_nmi_irqoff(vcpu);
 
 out:
 	guest_state_exit_irqoff();
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 35ad6dd5eaf6..96f3913f7fc5 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -13784,6 +13784,9 @@ int kvm_sev_es_string_io(struct kvm_vcpu *vcpu, unsigned int size,
 }
 EXPORT_SYMBOL_GPL(kvm_sev_es_string_io);
 
+EXPORT_SYMBOL_GPL(kvm_do_host_nmi_irqoff);
+EXPORT_SYMBOL_GPL(kvm_do_host_interrupt_irqoff);
+
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_entry);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_exit);
 EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_fast_mmio);
diff --git a/arch/x86/kvm/x86.h b/arch/x86/kvm/x86.h
index 5184fde1dc54..4d1430f8874b 100644
--- a/arch/x86/kvm/x86.h
+++ b/arch/x86/kvm/x86.h
@@ -491,6 +491,24 @@ static inline void kvm_machine_check(void)
 #endif
 }
 
+void kvm_do_host_nmi_irqoff(void);
+void kvm_do_host_interrupt_irqoff(unsigned long entry);
+
+static __always_inline void kvm_do_nmi_irqoff(struct kvm_vcpu *vcpu)
+{
+	kvm_before_interrupt(vcpu, KVM_HANDLING_NMI);
+	kvm_do_host_nmi_irqoff();
+	kvm_after_interrupt(vcpu);
+}
+
+static inline void kvm_do_interrupt_irqoff(struct kvm_vcpu *vcpu,
+					   unsigned long entry)
+{
+	kvm_before_interrupt(vcpu, KVM_HANDLING_IRQ);
+	kvm_do_host_interrupt_irqoff(entry);
+	kvm_after_interrupt(vcpu);
+}
+
 void kvm_load_guest_xsave_state(struct kvm_vcpu *vcpu);
 void kvm_load_host_xsave_state(struct kvm_vcpu *vcpu);
 int kvm_spec_ctrl_test_value(u64 value);
-- 
2.19.1.6.gb485710b