[PATCH stable v5.4+ 1/3] x86/kvm: Teardown PV features on boot CPU as well

[PATCH stable v5.4+ 1/3] x86/kvm: Teardown PV features on boot CPU as well

[Krzysztof Kozlowski]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

commit 8b79feffeca28c5459458fe78676b081e87c93a4 upstream.

Various PV features (Async PF, PV EOI, steal time) work through memory
shared with hypervisor and when we restore from hibernation we must
properly teardown all these features to make sure hypervisor doesn't
write to stale locations after we jump to the previously hibernated kernel
(which can try to place anything there). For secondary CPUs the job is
already done by kvm_cpu_down_prepare(), register syscore ops to do
the same for boot CPU.

Krzysztof:
This fixes memory corruption visible after second resume from
hibernation:

  BUG: Bad page state in process dbus-daemon  pfn:18b01
  page:ffffea000062c040 refcount:0 mapcount:0 mapping:0000000000000000 index:0x1 compound_mapcount: -30591
  flags: 0xfffffc0078141(locked|error|workingset|writeback|head|mappedtodisk|reclaim)
  raw: 000fffffc0078141 dead0000000002d0 dead000000000100 0000000000000000
  raw: 0000000000000001 0000000000000000 00000000ffffffff 0000000000000000
  page dumped because: PAGE_FLAGS_CHECK_AT_PREP flag set
  bad because of flags: 0x78141(locked|error|workingset|writeback|head|mappedtodisk|reclaim)

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210414123544.1060604-3-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Andrea Righi <andrea.righi@canonical.com>
[krzysztof: Extend the commit message]
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
---

Backport to v5.4 seems reasonable. Might have sense to earlier versions,
but this was not tested/investigated.

 arch/x86/kernel/kvm.c | 32 ++++++++++++++++++++++++++++----
 1 file changed, 28 insertions(+), 4 deletions(-)

diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index e820568ed4d5..6b906a651fb1 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -24,6 +24,7 @@
 #include <linux/debugfs.h>
 #include <linux/nmi.h>
 #include <linux/swait.h>
+#include <linux/syscore_ops.h>
 #include <asm/timer.h>
 #include <asm/cpu.h>
 #include <asm/traps.h>
@@ -558,17 +559,21 @@ static void kvm_guest_cpu_offline(void)
 
 static int kvm_cpu_online(unsigned int cpu)
 {
-	local_irq_disable();
+	unsigned long flags;
+
+	local_irq_save(flags);
 	kvm_guest_cpu_init();
-	local_irq_enable();
+	local_irq_restore(flags);
 	return 0;
 }
 
 static int kvm_cpu_down_prepare(unsigned int cpu)
 {
-	local_irq_disable();
+	unsigned long flags;
+
+	local_irq_save(flags);
 	kvm_guest_cpu_offline();
-	local_irq_enable();
+	local_irq_restore(flags);
 	return 0;
 }
 #endif
@@ -606,6 +611,23 @@ static void kvm_flush_tlb_others(const struct cpumask *cpumask,
 	native_flush_tlb_others(flushmask, info);
 }
 
+static int kvm_suspend(void)
+{
+	kvm_guest_cpu_offline();
+
+	return 0;
+}
+
+static void kvm_resume(void)
+{
+	kvm_cpu_online(raw_smp_processor_id());
+}
+
+static struct syscore_ops kvm_syscore_ops = {
+	.suspend	= kvm_suspend,
+	.resume		= kvm_resume,
+};
+
 static void __init kvm_guest_init(void)
 {
 	int i;
@@ -649,6 +671,8 @@ static void __init kvm_guest_init(void)
 	kvm_guest_cpu_init();
 #endif
 
+	register_syscore_ops(&kvm_syscore_ops);
+
 	/*
 	 * Hard lockup detection is enabled by default. Disable it, as guests
 	 * can get false positives too easily, for example if the host is
-- 
2.27.0

[PATCH stable v5.4+ 2/3] x86/kvm: Disable kvmclock on all CPUs on shutdown

[Krzysztof Kozlowski]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

commit c02027b5742b5aa804ef08a4a9db433295533046 upstream.

Currenly, we disable kvmclock from machine_shutdown() hook and this
only happens for boot CPU. We need to disable it for all CPUs to
guard against memory corruption e.g. on restore from hibernate.

Note, writing '0' to kvmclock MSR doesn't clear memory location, it
just prevents hypervisor from updating the location so for the short
while after write and while CPU is still alive, the clock remains usable
and correct so we don't need to switch to some other clocksource.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210414123544.1060604-4-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Andrea Righi <andrea.righi@canonical.com>
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
---
 arch/x86/include/asm/kvm_para.h | 4 ++--
 arch/x86/kernel/kvm.c           | 1 +
 arch/x86/kernel/kvmclock.c      | 5 +----
 3 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/arch/x86/include/asm/kvm_para.h b/arch/x86/include/asm/kvm_para.h
index 9b4df6eaa11a..a617fd360023 100644
--- a/arch/x86/include/asm/kvm_para.h
+++ b/arch/x86/include/asm/kvm_para.h
@@ -6,8 +6,6 @@
 #include <asm/alternative.h>
 #include <uapi/asm/kvm_para.h>
 
-extern void kvmclock_init(void);
-
 #ifdef CONFIG_KVM_GUEST
 bool kvm_check_and_clear_guest_paused(void);
 #else
@@ -85,6 +83,8 @@ static inline long kvm_hypercall4(unsigned int nr, unsigned long p1,
 }
 
 #ifdef CONFIG_KVM_GUEST
+void kvmclock_init(void);
+void kvmclock_disable(void);
 bool kvm_para_available(void);
 unsigned int kvm_arch_para_features(void);
 unsigned int kvm_arch_para_hints(void);
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index 6b906a651fb1..f535ba7714f8 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -555,6 +555,7 @@ static void kvm_guest_cpu_offline(void)
 		wrmsrl(MSR_KVM_PV_EOI_EN, 0);
 	kvm_pv_disable_apf();
 	apf_task_wake_all();
+	kvmclock_disable();
 }
 
 static int kvm_cpu_online(unsigned int cpu)
diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c
index 904494b924c1..bd3962953f78 100644
--- a/arch/x86/kernel/kvmclock.c
+++ b/arch/x86/kernel/kvmclock.c
@@ -214,11 +214,9 @@ static void kvm_crash_shutdown(struct pt_regs *regs)
 }
 #endif
 
-static void kvm_shutdown(void)
+void kvmclock_disable(void)
 {
 	native_write_msr(msr_kvm_system_time, 0, 0);
-	kvm_disable_steal_time();
-	native_machine_shutdown();
 }
 
 static void __init kvmclock_init_mem(void)
@@ -346,7 +344,6 @@ void __init kvmclock_init(void)
 #endif
 	x86_platform.save_sched_clock_state = kvm_save_sched_clock_state;
 	x86_platform.restore_sched_clock_state = kvm_restore_sched_clock_state;
-	machine_ops.shutdown  = kvm_shutdown;
 #ifdef CONFIG_KEXEC_CORE
 	machine_ops.crash_shutdown  = kvm_crash_shutdown;
 #endif
-- 
2.27.0

[PATCH stable v5.4+ 3/3] x86/kvm: Disable all PV features on crash

[Krzysztof Kozlowski]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

commit 3d6b84132d2a57b5a74100f6923a8feb679ac2ce upstream.

Crash shutdown handler only disables kvmclock and steal time, other PV
features remain active so we risk corrupting memory or getting some
side-effects in kdump kernel. Move crash handler to kvm.c and unify
with CPU offline.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210414123544.1060604-5-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Andrea Righi <andrea.righi@canonical.com>
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
---
 arch/x86/include/asm/kvm_para.h |  5 ----
 arch/x86/kernel/kvm.c           | 44 ++++++++++++++++++++++++---------
 arch/x86/kernel/kvmclock.c      | 21 ----------------
 3 files changed, 32 insertions(+), 38 deletions(-)

diff --git a/arch/x86/include/asm/kvm_para.h b/arch/x86/include/asm/kvm_para.h
index a617fd360023..7ff8ad490a78 100644
--- a/arch/x86/include/asm/kvm_para.h
+++ b/arch/x86/include/asm/kvm_para.h
@@ -91,7 +91,6 @@ unsigned int kvm_arch_para_hints(void);
 void kvm_async_pf_task_wait(u32 token, int interrupt_kernel);
 void kvm_async_pf_task_wake(u32 token);
 u32 kvm_read_and_reset_pf_reason(void);
-extern void kvm_disable_steal_time(void);
 void do_async_page_fault(struct pt_regs *regs, unsigned long error_code, unsigned long address);
 
 #ifdef CONFIG_PARAVIRT_SPINLOCKS
@@ -126,10 +125,6 @@ static inline u32 kvm_read_and_reset_pf_reason(void)
 	return 0;
 }
 
-static inline void kvm_disable_steal_time(void)
-{
-	return;
-}
 #endif
 
 #endif /* _ASM_X86_KVM_PARA_H */
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index f535ba7714f8..bff25b8166b7 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -34,6 +34,7 @@
 #include <asm/apicdef.h>
 #include <asm/hypervisor.h>
 #include <asm/tlb.h>
+#include <asm/reboot.h>
 
 static int kvmapf = 1;
 
@@ -352,6 +353,14 @@ static void kvm_pv_disable_apf(void)
 	       smp_processor_id());
 }
 
+static void kvm_disable_steal_time(void)
+{
+	if (!has_steal_clock)
+		return;
+
+	wrmsr(MSR_KVM_STEAL_TIME, 0, 0);
+}
+
 static void kvm_pv_guest_cpu_reboot(void *unused)
 {
 	/*
@@ -394,14 +403,6 @@ static u64 kvm_steal_clock(int cpu)
 	return steal;
 }
 
-void kvm_disable_steal_time(void)
-{
-	if (!has_steal_clock)
-		return;
-
-	wrmsr(MSR_KVM_STEAL_TIME, 0, 0);
-}
-
 static inline void __set_percpu_decrypted(void *ptr, unsigned long size)
 {
 	early_set_memory_decrypted((unsigned long) ptr, size);
@@ -548,13 +549,14 @@ static void __init kvm_smp_prepare_boot_cpu(void)
 	kvm_spinlock_init();
 }
 
-static void kvm_guest_cpu_offline(void)
+static void kvm_guest_cpu_offline(bool shutdown)
 {
 	kvm_disable_steal_time();
 	if (kvm_para_has_feature(KVM_FEATURE_PV_EOI))
 		wrmsrl(MSR_KVM_PV_EOI_EN, 0);
 	kvm_pv_disable_apf();
-	apf_task_wake_all();
+	if (!shutdown)
+		apf_task_wake_all();
 	kvmclock_disable();
 }
 
@@ -573,7 +575,7 @@ static int kvm_cpu_down_prepare(unsigned int cpu)
 	unsigned long flags;
 
 	local_irq_save(flags);
-	kvm_guest_cpu_offline();
+	kvm_guest_cpu_offline(false);
 	local_irq_restore(flags);
 	return 0;
 }
@@ -614,7 +616,7 @@ static void kvm_flush_tlb_others(const struct cpumask *cpumask,
 
 static int kvm_suspend(void)
 {
-	kvm_guest_cpu_offline();
+	kvm_guest_cpu_offline(false);
 
 	return 0;
 }
@@ -629,6 +631,20 @@ static struct syscore_ops kvm_syscore_ops = {
 	.resume		= kvm_resume,
 };
 
+/*
+ * After a PV feature is registered, the host will keep writing to the
+ * registered memory location. If the guest happens to shutdown, this memory
+ * won't be valid. In cases like kexec, in which you install a new kernel, this
+ * means a random memory location will be kept being written.
+ */
+#ifdef CONFIG_KEXEC_CORE
+static void kvm_crash_shutdown(struct pt_regs *regs)
+{
+	kvm_guest_cpu_offline(true);
+	native_machine_crash_shutdown(regs);
+}
+#endif
+
 static void __init kvm_guest_init(void)
 {
 	int i;
@@ -672,6 +688,10 @@ static void __init kvm_guest_init(void)
 	kvm_guest_cpu_init();
 #endif
 
+#ifdef CONFIG_KEXEC_CORE
+	machine_ops.crash_shutdown = kvm_crash_shutdown;
+#endif
+
 	register_syscore_ops(&kvm_syscore_ops);
 
 	/*
diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c
index bd3962953f78..4a0802af2e3e 100644
--- a/arch/x86/kernel/kvmclock.c
+++ b/arch/x86/kernel/kvmclock.c
@@ -20,7 +20,6 @@
 #include <asm/hypervisor.h>
 #include <asm/mem_encrypt.h>
 #include <asm/x86_init.h>
-#include <asm/reboot.h>
 #include <asm/kvmclock.h>
 
 static int kvmclock __initdata = 1;
@@ -197,23 +196,6 @@ static void kvm_setup_secondary_clock(void)
 }
 #endif
 
-/*
- * After the clock is registered, the host will keep writing to the
- * registered memory location. If the guest happens to shutdown, this memory
- * won't be valid. In cases like kexec, in which you install a new kernel, this
- * means a random memory location will be kept being written. So before any
- * kind of shutdown from our side, we unregister the clock by writing anything
- * that does not have the 'enable' bit set in the msr
- */
-#ifdef CONFIG_KEXEC_CORE
-static void kvm_crash_shutdown(struct pt_regs *regs)
-{
-	native_write_msr(msr_kvm_system_time, 0, 0);
-	kvm_disable_steal_time();
-	native_machine_crash_shutdown(regs);
-}
-#endif
-
 void kvmclock_disable(void)
 {
 	native_write_msr(msr_kvm_system_time, 0, 0);
@@ -344,9 +326,6 @@ void __init kvmclock_init(void)
 #endif
 	x86_platform.save_sched_clock_state = kvm_save_sched_clock_state;
 	x86_platform.restore_sched_clock_state = kvm_restore_sched_clock_state;
-#ifdef CONFIG_KEXEC_CORE
-	machine_ops.crash_shutdown  = kvm_crash_shutdown;
-#endif
 	kvm_get_preset_lpj();
 
 	/*
-- 
2.27.0

Re: [PATCH stable v5.4+ 1/3] x86/kvm: Teardown PV features on boot CPU as well

[Paolo Bonzini]

On 20/05/21 14:56, Krzysztof Kozlowski wrote:
> From: Vitaly Kuznetsov <vkuznets@redhat.com>
> 
> commit 8b79feffeca28c5459458fe78676b081e87c93a4 upstream.
> 
> Various PV features (Async PF, PV EOI, steal time) work through memory
> shared with hypervisor and when we restore from hibernation we must
> properly teardown all these features to make sure hypervisor doesn't
> write to stale locations after we jump to the previously hibernated kernel
> (which can try to place anything there). For secondary CPUs the job is
> already done by kvm_cpu_down_prepare(), register syscore ops to do
> the same for boot CPU.
> 
> Krzysztof:
> This fixes memory corruption visible after second resume from
> hibernation:

Hi, you should include a cover letter detailing the differences between 
the original patches and the backport.

(I'll review it anyway, but it would have helped).

Paolo

>    BUG: Bad page state in process dbus-daemon  pfn:18b01
>    page:ffffea000062c040 refcount:0 mapcount:0 mapping:0000000000000000 index:0x1 compound_mapcount: -30591
>    flags: 0xfffffc0078141(locked|error|workingset|writeback|head|mappedtodisk|reclaim)
>    raw: 000fffffc0078141 dead0000000002d0 dead000000000100 0000000000000000
>    raw: 0000000000000001 0000000000000000 00000000ffffffff 0000000000000000
>    page dumped because: PAGE_FLAGS_CHECK_AT_PREP flag set
>    bad because of flags: 0x78141(locked|error|workingset|writeback|head|mappedtodisk|reclaim)
> 
> Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
> Message-Id: <20210414123544.1060604-3-vkuznets@redhat.com>
> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
> Signed-off-by: Andrea Righi <andrea.righi@canonical.com>
> [krzysztof: Extend the commit message]
> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
> ---
> 
> Backport to v5.4 seems reasonable. Might have sense to earlier versions,
> but this was not tested/investigated.
> 
>   arch/x86/kernel/kvm.c | 32 ++++++++++++++++++++++++++++----
>   1 file changed, 28 insertions(+), 4 deletions(-)
> 
> diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
> index e820568ed4d5..6b906a651fb1 100644
> --- a/arch/x86/kernel/kvm.c
> +++ b/arch/x86/kernel/kvm.c
> @@ -24,6 +24,7 @@
>   #include <linux/debugfs.h>
>   #include <linux/nmi.h>
>   #include <linux/swait.h>
> +#include <linux/syscore_ops.h>
>   #include <asm/timer.h>
>   #include <asm/cpu.h>
>   #include <asm/traps.h>
> @@ -558,17 +559,21 @@ static void kvm_guest_cpu_offline(void)
>   
>   static int kvm_cpu_online(unsigned int cpu)
>   {
> -	local_irq_disable();
> +	unsigned long flags;
> +
> +	local_irq_save(flags);
>   	kvm_guest_cpu_init();
> -	local_irq_enable();
> +	local_irq_restore(flags);
>   	return 0;
>   }
>   
>   static int kvm_cpu_down_prepare(unsigned int cpu)
>   {
> -	local_irq_disable();
> +	unsigned long flags;
> +
> +	local_irq_save(flags);
>   	kvm_guest_cpu_offline();
> -	local_irq_enable();
> +	local_irq_restore(flags);
>   	return 0;
>   }
>   #endif
> @@ -606,6 +611,23 @@ static void kvm_flush_tlb_others(const struct cpumask *cpumask,
>   	native_flush_tlb_others(flushmask, info);
>   }
>   
> +static int kvm_suspend(void)
> +{
> +	kvm_guest_cpu_offline();
> +
> +	return 0;
> +}
> +
> +static void kvm_resume(void)
> +{
> +	kvm_cpu_online(raw_smp_processor_id());
> +}
> +
> +static struct syscore_ops kvm_syscore_ops = {
> +	.suspend	= kvm_suspend,
> +	.resume		= kvm_resume,
> +};
> +
>   static void __init kvm_guest_init(void)
>   {
>   	int i;
> @@ -649,6 +671,8 @@ static void __init kvm_guest_init(void)
>   	kvm_guest_cpu_init();
>   #endif
>   
> +	register_syscore_ops(&kvm_syscore_ops);
> +
>   	/*
>   	 * Hard lockup detection is enabled by default. Disable it, as guests
>   	 * can get false positives too easily, for example if the host is
> 

Re: [PATCH stable v5.4+ 1/3] x86/kvm: Teardown PV features on boot CPU as well

[Krzysztof Kozlowski]

On 20/05/2021 12:47, Paolo Bonzini wrote:
> On 20/05/21 14:56, Krzysztof Kozlowski wrote:
>> From: Vitaly Kuznetsov <vkuznets@redhat.com>
>>
>> commit 8b79feffeca28c5459458fe78676b081e87c93a4 upstream.
>>
>> Various PV features (Async PF, PV EOI, steal time) work through memory
>> shared with hypervisor and when we restore from hibernation we must
>> properly teardown all these features to make sure hypervisor doesn't
>> write to stale locations after we jump to the previously hibernated kernel
>> (which can try to place anything there). For secondary CPUs the job is
>> already done by kvm_cpu_down_prepare(), register syscore ops to do
>> the same for boot CPU.
>>
>> Krzysztof:
>> This fixes memory corruption visible after second resume from
>> hibernation:
> 
> Hi, you should include a cover letter detailing the differences between 
> the original patches and the backport.
> 
> (I'll review it anyway, but it would have helped).

My bad, I actually was not aware that backport differs that much. I can
describe in v2.

The patch context looks quite a different and now I see
kvm_guest_cpu_offline() ends up within CONFIG_SMP for unclear reasons.
Let me try to fix it in v2.


Best regards,
Krzysztof