diff mbox series

KVM: x86: hyper-v: Task srcu lock when accessing kvm_memslots()

Message ID 1620634919-4563-1-git-send-email-wanpengli@tencent.com (mailing list archive)
State New, archived
Headers show
Series KVM: x86: hyper-v: Task srcu lock when accessing kvm_memslots() | expand

Commit Message

Wanpeng Li May 10, 2021, 8:21 a.m. UTC 
From: Wanpeng Li <wanpengli@tencent.com>

 WARNING: suspicious RCU usage
 5.13.0-rc1 #4 Not tainted
 -----------------------------
 ./include/linux/kvm_host.h:710 suspicious rcu_dereference_check() usage!
 
other info that might help us debug this:

rcu_scheduler_active = 2, debug_locks = 1
 1 lock held by hyperv_clock/8318:
  #0: ffffb6b8cb05a7d8 (&hv->hv_lock){+.+.}-{3:3}, at: kvm_hv_invalidate_tsc_page+0x3e/0xa0 [kvm]
 
stack backtrace:
CPU: 3 PID: 8318 Comm: hyperv_clock Not tainted 5.13.0-rc1 #4
Call Trace:
 dump_stack+0x87/0xb7
 lockdep_rcu_suspicious+0xce/0xf0
 kvm_write_guest_page+0x1c1/0x1d0 [kvm]
 kvm_write_guest+0x50/0x90 [kvm]
 kvm_hv_invalidate_tsc_page+0x79/0xa0 [kvm]
 kvm_gen_update_masterclock+0x1d/0x110 [kvm]
 kvm_arch_vm_ioctl+0x2a7/0xc50 [kvm]
 kvm_vm_ioctl+0x123/0x11d0 [kvm]
 __x64_sys_ioctl+0x3ed/0x9d0
 do_syscall_64+0x3d/0x80
 entry_SYSCALL_64_after_hwframe+0x44/0xae

kvm_memslots() will be called by kvm_write_guest(), so we should take the srcu lock.

Fixes: e880c6ea5 (KVM: x86: hyper-v: Prevent using not-yet-updated TSC page by secondary CPUs)
Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
---
 arch/x86/kvm/hyperv.c | 8 ++++++++
 1 file changed, 8 insertions(+)

Comments

Vitaly Kuznetsov May 10, 2021, 8:48 a.m. UTC | #1 
Wanpeng Li <kernellwp@gmail.com> writes:

> From: Wanpeng Li <wanpengli@tencent.com>
>
>  WARNING: suspicious RCU usage
>  5.13.0-rc1 #4 Not tainted
>  -----------------------------
>  ./include/linux/kvm_host.h:710 suspicious rcu_dereference_check() usage!
>  
> other info that might help us debug this:
>
> rcu_scheduler_active = 2, debug_locks = 1
>  1 lock held by hyperv_clock/8318:
>   #0: ffffb6b8cb05a7d8 (&hv->hv_lock){+.+.}-{3:3}, at: kvm_hv_invalidate_tsc_page+0x3e/0xa0 [kvm]
>  
> stack backtrace:
> CPU: 3 PID: 8318 Comm: hyperv_clock Not tainted 5.13.0-rc1 #4
> Call Trace:
>  dump_stack+0x87/0xb7
>  lockdep_rcu_suspicious+0xce/0xf0
>  kvm_write_guest_page+0x1c1/0x1d0 [kvm]
>  kvm_write_guest+0x50/0x90 [kvm]
>  kvm_hv_invalidate_tsc_page+0x79/0xa0 [kvm]
>  kvm_gen_update_masterclock+0x1d/0x110 [kvm]
>  kvm_arch_vm_ioctl+0x2a7/0xc50 [kvm]
>  kvm_vm_ioctl+0x123/0x11d0 [kvm]
>  __x64_sys_ioctl+0x3ed/0x9d0
>  do_syscall_64+0x3d/0x80
>  entry_SYSCALL_64_after_hwframe+0x44/0xae
>
> kvm_memslots() will be called by kvm_write_guest(), so we should take the srcu lock.
>
> Fixes: e880c6ea5 (KVM: x86: hyper-v: Prevent using not-yet-updated TSC page by secondary CPUs)
> Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
> ---
>  arch/x86/kvm/hyperv.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
>
> diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
> index f98370a3..f00830e 100644
> --- a/arch/x86/kvm/hyperv.c
> +++ b/arch/x86/kvm/hyperv.c
> @@ -1172,6 +1172,7 @@ void kvm_hv_invalidate_tsc_page(struct kvm *kvm)
>  {
>  	struct kvm_hv *hv = to_kvm_hv(kvm);
>  	u64 gfn;
> +	int idx;
>  
>  	if (hv->hv_tsc_page_status == HV_TSC_PAGE_BROKEN ||
>  	    hv->hv_tsc_page_status == HV_TSC_PAGE_UNSET ||
> @@ -1190,9 +1191,16 @@ void kvm_hv_invalidate_tsc_page(struct kvm *kvm)
>  	gfn = hv->hv_tsc_page >> HV_X64_MSR_TSC_REFERENCE_ADDRESS_SHIFT;
>  
>  	hv->tsc_ref.tsc_sequence = 0;
> +
> +	/*
> +	 * Take the srcu lock as memslots will be accessed to check the gfn
> +	 * cache generation against the memslots generation.
> +	 */
> +	idx = srcu_read_lock(&kvm->srcu);
>  	if (kvm_write_guest(kvm, gfn_to_gpa(gfn),
>  			    &hv->tsc_ref, sizeof(hv->tsc_ref.tsc_sequence)))
>  		hv->hv_tsc_page_status = HV_TSC_PAGE_BROKEN;
> +	srcu_read_unlock(&kvm->srcu, idx);
>  
>  out_unlock:
>  	mutex_unlock(&hv->hv_lock);

Thanks! 

Do we need to do the same in kvm_hv_setup_tsc_page()?
Wanpeng Li May 10, 2021, 8:51 a.m. UTC | #2 
On Mon, 10 May 2021 at 16:48, Vitaly Kuznetsov <vkuznets@redhat.com> wrote:
>
> Wanpeng Li <kernellwp@gmail.com> writes:
>
> > From: Wanpeng Li <wanpengli@tencent.com>
> >
> >  WARNING: suspicious RCU usage
> >  5.13.0-rc1 #4 Not tainted
> >  -----------------------------
> >  ./include/linux/kvm_host.h:710 suspicious rcu_dereference_check() usage!
> >
> > other info that might help us debug this:
> >
> > rcu_scheduler_active = 2, debug_locks = 1
> >  1 lock held by hyperv_clock/8318:
> >   #0: ffffb6b8cb05a7d8 (&hv->hv_lock){+.+.}-{3:3}, at: kvm_hv_invalidate_tsc_page+0x3e/0xa0 [kvm]
> >
> > stack backtrace:
> > CPU: 3 PID: 8318 Comm: hyperv_clock Not tainted 5.13.0-rc1 #4
> > Call Trace:
> >  dump_stack+0x87/0xb7
> >  lockdep_rcu_suspicious+0xce/0xf0
> >  kvm_write_guest_page+0x1c1/0x1d0 [kvm]
> >  kvm_write_guest+0x50/0x90 [kvm]
> >  kvm_hv_invalidate_tsc_page+0x79/0xa0 [kvm]
> >  kvm_gen_update_masterclock+0x1d/0x110 [kvm]
> >  kvm_arch_vm_ioctl+0x2a7/0xc50 [kvm]
> >  kvm_vm_ioctl+0x123/0x11d0 [kvm]
> >  __x64_sys_ioctl+0x3ed/0x9d0
> >  do_syscall_64+0x3d/0x80
> >  entry_SYSCALL_64_after_hwframe+0x44/0xae
> >
> > kvm_memslots() will be called by kvm_write_guest(), so we should take the srcu lock.
> >
> > Fixes: e880c6ea5 (KVM: x86: hyper-v: Prevent using not-yet-updated TSC page by secondary CPUs)
> > Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
> > ---
> >  arch/x86/kvm/hyperv.c | 8 ++++++++
> >  1 file changed, 8 insertions(+)
> >
> > diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
> > index f98370a3..f00830e 100644
> > --- a/arch/x86/kvm/hyperv.c
> > +++ b/arch/x86/kvm/hyperv.c
> > @@ -1172,6 +1172,7 @@ void kvm_hv_invalidate_tsc_page(struct kvm *kvm)
> >  {
> >       struct kvm_hv *hv = to_kvm_hv(kvm);
> >       u64 gfn;
> > +     int idx;
> >
> >       if (hv->hv_tsc_page_status == HV_TSC_PAGE_BROKEN ||
> >           hv->hv_tsc_page_status == HV_TSC_PAGE_UNSET ||
> > @@ -1190,9 +1191,16 @@ void kvm_hv_invalidate_tsc_page(struct kvm *kvm)
> >       gfn = hv->hv_tsc_page >> HV_X64_MSR_TSC_REFERENCE_ADDRESS_SHIFT;
> >
> >       hv->tsc_ref.tsc_sequence = 0;
> > +
> > +     /*
> > +      * Take the srcu lock as memslots will be accessed to check the gfn
> > +      * cache generation against the memslots generation.
> > +      */
> > +     idx = srcu_read_lock(&kvm->srcu);
> >       if (kvm_write_guest(kvm, gfn_to_gpa(gfn),
> >                           &hv->tsc_ref, sizeof(hv->tsc_ref.tsc_sequence)))
> >               hv->hv_tsc_page_status = HV_TSC_PAGE_BROKEN;
> > +     srcu_read_unlock(&kvm->srcu, idx);
> >
> >  out_unlock:
> >       mutex_unlock(&hv->hv_lock);
>
> Thanks!
>
> Do we need to do the same in kvm_hv_setup_tsc_page()?

kvm_hv_setup_tsc_page() is called in vcpu_enter_guest() path which has
already held kvm->srcu lock.

    Wanpeng
Vitaly Kuznetsov May 10, 2021, 11:15 a.m. UTC | #3 
Wanpeng Li <kernellwp@gmail.com> writes:

> On Mon, 10 May 2021 at 16:48, Vitaly Kuznetsov <vkuznets@redhat.com> wrote:
>>
>> Wanpeng Li <kernellwp@gmail.com> writes:
>>
>> > From: Wanpeng Li <wanpengli@tencent.com>
>> >
>> >  WARNING: suspicious RCU usage
>> >  5.13.0-rc1 #4 Not tainted
>> >  -----------------------------
>> >  ./include/linux/kvm_host.h:710 suspicious rcu_dereference_check() usage!
>> >
>> > other info that might help us debug this:
>> >
>> > rcu_scheduler_active = 2, debug_locks = 1
>> >  1 lock held by hyperv_clock/8318:
>> >   #0: ffffb6b8cb05a7d8 (&hv->hv_lock){+.+.}-{3:3}, at: kvm_hv_invalidate_tsc_page+0x3e/0xa0 [kvm]
>> >
>> > stack backtrace:
>> > CPU: 3 PID: 8318 Comm: hyperv_clock Not tainted 5.13.0-rc1 #4
>> > Call Trace:
>> >  dump_stack+0x87/0xb7
>> >  lockdep_rcu_suspicious+0xce/0xf0
>> >  kvm_write_guest_page+0x1c1/0x1d0 [kvm]
>> >  kvm_write_guest+0x50/0x90 [kvm]
>> >  kvm_hv_invalidate_tsc_page+0x79/0xa0 [kvm]
>> >  kvm_gen_update_masterclock+0x1d/0x110 [kvm]
>> >  kvm_arch_vm_ioctl+0x2a7/0xc50 [kvm]
>> >  kvm_vm_ioctl+0x123/0x11d0 [kvm]
>> >  __x64_sys_ioctl+0x3ed/0x9d0
>> >  do_syscall_64+0x3d/0x80
>> >  entry_SYSCALL_64_after_hwframe+0x44/0xae
>> >
>> > kvm_memslots() will be called by kvm_write_guest(), so we should take the srcu lock.
>> >
>> > Fixes: e880c6ea5 (KVM: x86: hyper-v: Prevent using not-yet-updated TSC page by secondary CPUs)
>> > Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
>> > ---
>> >  arch/x86/kvm/hyperv.c | 8 ++++++++
>> >  1 file changed, 8 insertions(+)
>> >
>> > diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
>> > index f98370a3..f00830e 100644
>> > --- a/arch/x86/kvm/hyperv.c
>> > +++ b/arch/x86/kvm/hyperv.c
>> > @@ -1172,6 +1172,7 @@ void kvm_hv_invalidate_tsc_page(struct kvm *kvm)
>> >  {
>> >       struct kvm_hv *hv = to_kvm_hv(kvm);
>> >       u64 gfn;
>> > +     int idx;
>> >
>> >       if (hv->hv_tsc_page_status == HV_TSC_PAGE_BROKEN ||
>> >           hv->hv_tsc_page_status == HV_TSC_PAGE_UNSET ||
>> > @@ -1190,9 +1191,16 @@ void kvm_hv_invalidate_tsc_page(struct kvm *kvm)
>> >       gfn = hv->hv_tsc_page >> HV_X64_MSR_TSC_REFERENCE_ADDRESS_SHIFT;
>> >
>> >       hv->tsc_ref.tsc_sequence = 0;
>> > +
>> > +     /*
>> > +      * Take the srcu lock as memslots will be accessed to check the gfn
>> > +      * cache generation against the memslots generation.
>> > +      */
>> > +     idx = srcu_read_lock(&kvm->srcu);
>> >       if (kvm_write_guest(kvm, gfn_to_gpa(gfn),
>> >                           &hv->tsc_ref, sizeof(hv->tsc_ref.tsc_sequence)))
>> >               hv->hv_tsc_page_status = HV_TSC_PAGE_BROKEN;
>> > +     srcu_read_unlock(&kvm->srcu, idx);
>> >
>> >  out_unlock:
>> >       mutex_unlock(&hv->hv_lock);
>>
>> Thanks!
>>
>> Do we need to do the same in kvm_hv_setup_tsc_page()?
>
> kvm_hv_setup_tsc_page() is called in vcpu_enter_guest() path which has
> already held kvm->srcu lock.
>

I, true, thanks for checking!

For the patch:
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
diff mbox series

Patch

diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
index f98370a3..f00830e 100644
--- a/arch/x86/kvm/hyperv.c
+++ b/arch/x86/kvm/hyperv.c
@@ -1172,6 +1172,7 @@  void kvm_hv_invalidate_tsc_page(struct kvm *kvm)
 {
 	struct kvm_hv *hv = to_kvm_hv(kvm);
 	u64 gfn;
+	int idx;
 
 	if (hv->hv_tsc_page_status == HV_TSC_PAGE_BROKEN ||
 	    hv->hv_tsc_page_status == HV_TSC_PAGE_UNSET ||
@@ -1190,9 +1191,16 @@  void kvm_hv_invalidate_tsc_page(struct kvm *kvm)
 	gfn = hv->hv_tsc_page >> HV_X64_MSR_TSC_REFERENCE_ADDRESS_SHIFT;
 
 	hv->tsc_ref.tsc_sequence = 0;
+
+	/*
+	 * Take the srcu lock as memslots will be accessed to check the gfn
+	 * cache generation against the memslots generation.
+	 */
+	idx = srcu_read_lock(&kvm->srcu);
 	if (kvm_write_guest(kvm, gfn_to_gpa(gfn),
 			    &hv->tsc_ref, sizeof(hv->tsc_ref.tsc_sequence)))
 		hv->hv_tsc_page_status = HV_TSC_PAGE_BROKEN;
+	srcu_read_unlock(&kvm->srcu, idx);
 
 out_unlock:
 	mutex_unlock(&hv->hv_lock);