linux-kernel - [PATCH 4/7] KVM: VMX: Always inline to_vmx() and to_kvm

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

Message-ID: <20221213060912.654668-5-seanjc@google.com>
Date:   Tue, 13 Dec 2022 06:09:09 +0000
From:   Sean Christopherson <seanjc@...gle.com>
To:     Sean Christopherson <seanjc@...gle.com>,
        Paolo Bonzini <pbonzini@...hat.com>
Cc:     kvm@...r.kernel.org, linux-kernel@...r.kernel.org,
        Peter Zijlstra <peterz@...radead.org>,
        Andy Lutomirski <luto@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>
Subject: [PATCH 4/7] KVM: VMX: Always inline to_vmx() and to_kvm_vmx()

Tag to_vmx() and to_kvm_vmx() __always_inline as they both just reflect
the passed in pointer (the embedded struct is the first field in the
container), and drop the @vmx param from vmx_vcpu_enter_exit(), which
likely existed purely to make noinstr validation happy.

Amusingly, when the compiler decides to not inline the helpers, e.g. for
KASAN builds, to_vmx() and to_kvm_vmx() may end up pointing at the same
symbol, which generates very confusing objtool warnings.  E.g. the use of
to_vmx() in a future patch led to objtool complaining about to_kvm_vmx(),
and only once all use of to_kvm_vmx() was commented out did to_vmx() pop
up in the obj tool report.

  vmlinux.o: warning: objtool: vmx_vcpu_enter_exit+0x160: call to to_kvm_vmx()
                               leaves .noinstr.text section

Signed-off-by: Sean Christopherson <seanjc@...gle.com>
---
 arch/x86/kvm/vmx/vmx.c | 5 +++--
 arch/x86/kvm/vmx/vmx.h | 4 ++--
 2 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index fe5615fd8295..e2c96f204b82 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -7096,9 +7096,10 @@ static fastpath_t vmx_exit_handlers_fastpath(struct kvm_vcpu *vcpu)
 }
 
 static noinstr void vmx_vcpu_enter_exit(struct kvm_vcpu *vcpu,
-					struct vcpu_vmx *vmx,
 					unsigned long flags)
 {
+	struct vcpu_vmx *vmx = to_vmx(vcpu);
+
 	guest_state_enter_irqoff();
 
 	/* L1D Flush includes CPU buffer clear to mitigate MDS */
@@ -7216,7 +7217,7 @@ static fastpath_t vmx_vcpu_run(struct kvm_vcpu *vcpu)
 	kvm_wait_lapic_expire(vcpu);
 
 	/* The actual VMENTER/EXIT is in the .noinstr.text section. */
-	vmx_vcpu_enter_exit(vcpu, vmx, __vmx_vcpu_run_flags(vmx));
+	vmx_vcpu_enter_exit(vcpu, __vmx_vcpu_run_flags(vmx));
 
 	/* All fields are clean at this point */
 	if (static_branch_unlikely(&enable_evmcs)) {
diff --git a/arch/x86/kvm/vmx/vmx.h b/arch/x86/kvm/vmx/vmx.h
index bb720a2f11ab..2acdc54bc34b 100644
--- a/arch/x86/kvm/vmx/vmx.h
+++ b/arch/x86/kvm/vmx/vmx.h
@@ -640,12 +640,12 @@ BUILD_CONTROLS_SHADOW(tertiary_exec, TERTIARY_VM_EXEC_CONTROL, 64)
 				(1 << VCPU_EXREG_EXIT_INFO_1) | \
 				(1 << VCPU_EXREG_EXIT_INFO_2))
 
-static inline struct kvm_vmx *to_kvm_vmx(struct kvm *kvm)
+static __always_inline struct kvm_vmx *to_kvm_vmx(struct kvm *kvm)
 {
 	return container_of(kvm, struct kvm_vmx, kvm);
 }
 
-static inline struct vcpu_vmx *to_vmx(struct kvm_vcpu *vcpu)
+static __always_inline struct vcpu_vmx *to_vmx(struct kvm_vcpu *vcpu)
 {
 	return container_of(vcpu, struct vcpu_vmx, vcpu);
 }
-- 
2.39.0.rc1.256.g54fd8350bd-goog