linux-kernel - [RFC KVM 08/27] KVM: x86: Optimize branches which checks if address space isolation enabled

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: <1557758315-12667-9-git-send-email-alexandre.chartre@oracle.com>
Date:   Mon, 13 May 2019 16:38:16 +0200
From:   Alexandre Chartre <alexandre.chartre@...cle.com>
To:     pbonzini@...hat.com, rkrcmar@...hat.com, tglx@...utronix.de,
        mingo@...hat.com, bp@...en8.de, hpa@...or.com,
        dave.hansen@...ux.intel.com, luto@...nel.org, peterz@...radead.org,
        kvm@...r.kernel.org, x86@...nel.org, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Cc:     konrad.wilk@...cle.com, jan.setjeeilers@...cle.com,
        liran.alon@...cle.com, jwadams@...gle.com,
        alexandre.chartre@...cle.com
Subject: [RFC KVM 08/27] KVM: x86: Optimize branches which checks if address space isolation enabled

From: Liran Alon <liran.alon@...cle.com>

As every entry to guest checks if should switch from host_mm to kvm_mm,
these branches is at very hot path. Optimize them by using
static_branch.

Signed-off-by: Liran Alon <liran.alon@...cle.com>
Signed-off-by: Alexandre Chartre <alexandre.chartre@...cle.com>
---
 arch/x86/kvm/isolation.c |   11 ++++++++---
 arch/x86/kvm/isolation.h |    7 +++++++
 2 files changed, 15 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kvm/isolation.c b/arch/x86/kvm/isolation.c
index eeb60c4..43fd924 100644
--- a/arch/x86/kvm/isolation.c
+++ b/arch/x86/kvm/isolation.c
@@ -23,6 +23,9 @@ struct mm_struct kvm_mm = {
 	.mmlist			= LIST_HEAD_INIT(kvm_mm.mmlist),
 };
 
+DEFINE_STATIC_KEY_FALSE(kvm_isolation_enabled);
+EXPORT_SYMBOL(kvm_isolation_enabled);
+
 /*
  * When set to true, KVM #VMExit handlers run in isolated address space
  * which maps only KVM required code and per-VM information instead of
@@ -118,15 +121,17 @@ int kvm_isolation_init(void)
 
 	kvm_isolation_set_handlers();
 	pr_info("KVM: x86: Running with isolated address space\n");
+	static_branch_enable(&kvm_isolation_enabled);
 
 	return 0;
 }
 
 void kvm_isolation_uninit(void)
 {
-	if (!address_space_isolation)
+	if (!kvm_isolation())
 		return;
 
+	static_branch_disable(&kvm_isolation_enabled);
 	kvm_isolation_clear_handlers();
 	kvm_isolation_uninit_mm();
 	pr_info("KVM: x86: End of isolated address space\n");
@@ -140,7 +145,7 @@ void kvm_may_access_sensitive_data(struct kvm_vcpu *vcpu)
 
 void kvm_isolation_enter(void)
 {
-	if (address_space_isolation) {
+	if (kvm_isolation()) {
 		/*
 		 * Switches to kvm_mm should happen from vCPU thread,
 		 * which should not be a kernel thread with no mm
@@ -152,7 +157,7 @@ void kvm_isolation_enter(void)
 
 void kvm_isolation_exit(void)
 {
-	if (address_space_isolation) {
+	if (kvm_isolation()) {
 		/* TODO: Kick sibling hyperthread before switch to host mm */
 		/* TODO: switch back to original mm */
 	}
diff --git a/arch/x86/kvm/isolation.h b/arch/x86/kvm/isolation.h
index 1290d32..aa5e979 100644
--- a/arch/x86/kvm/isolation.h
+++ b/arch/x86/kvm/isolation.h
@@ -4,6 +4,13 @@
 
 #include <linux/kvm_host.h>
 
+DECLARE_STATIC_KEY_FALSE(kvm_isolation_enabled);
+
+static inline bool kvm_isolation(void)
+{
+	return static_branch_likely(&kvm_isolation_enabled);
+}
+
 extern int kvm_isolation_init(void);
 extern void kvm_isolation_uninit(void);
 extern void kvm_isolation_enter(void);
-- 
1.7.1