linux-kernel - Re: [PATCH] KVM: x86/xen: Take srcu lock when accessing 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: <f025b59c-5a8a-abf7-20fc-323a5b450ba5@redhat.com>
Date:   Fri, 23 Apr 2021 10:13:05 +0200
From:   Paolo Bonzini <pbonzini@...hat.com>
To:     Wanpeng Li <kernellwp@...il.com>, linux-kernel@...r.kernel.org,
        kvm@...r.kernel.org
Cc:     Sean Christopherson <seanjc@...gle.com>,
        Vitaly Kuznetsov <vkuznets@...hat.com>,
        Wanpeng Li <wanpengli@...cent.com>,
        Jim Mattson <jmattson@...gle.com>,
        Joerg Roedel <joro@...tes.org>
Subject: Re: [PATCH] KVM: x86/xen: Take srcu lock when accessing
 kvm_memslots()

On 23/04/21 09:11, Wanpeng Li wrote:
> From: Wanpeng Li <wanpengli@...cent.com>
> 
> kvm_memslots() will be called by kvm_write_guest_offset_cached() so
> take the srcu lock.
> 
> Signed-off-by: Wanpeng Li <wanpengli@...cent.com>

Good catch.  But I would pull it from kvm_steal_time_set_preempted to 
kvm_arch_vcpu_put instead.

Paolo

> ---
>   arch/x86/kvm/xen.c | 18 ++++++++++++++----
>   1 file changed, 14 insertions(+), 4 deletions(-)
> 
> diff --git a/arch/x86/kvm/xen.c b/arch/x86/kvm/xen.c
> index ae17250..d0df782 100644
> --- a/arch/x86/kvm/xen.c
> +++ b/arch/x86/kvm/xen.c
> @@ -96,6 +96,7 @@ void kvm_xen_update_runstate_guest(struct kvm_vcpu *v, int state)
>   	struct kvm_vcpu_xen *vx = &v->arch.xen;
>   	uint64_t state_entry_time;
>   	unsigned int offset;
> +	int idx;
>   
>   	kvm_xen_update_runstate(v, state);
>   
> @@ -133,10 +134,16 @@ void kvm_xen_update_runstate_guest(struct kvm_vcpu *v, int state)
>   	BUILD_BUG_ON(sizeof(((struct compat_vcpu_runstate_info *)0)->state_entry_time) !=
>   		     sizeof(state_entry_time));
>   
> +	/*
> +	 * Take the srcu lock as memslots will be accessed to check the gfn
> +	 * cache generation against the memslots generation.
> +	 */
> +	idx = srcu_read_lock(&v->kvm->srcu);
> +
>   	if (kvm_write_guest_offset_cached(v->kvm, &v->arch.xen.runstate_cache,
>   					  &state_entry_time, offset,
>   					  sizeof(state_entry_time)))
> -		return;
> +		goto out;
>   	smp_wmb();
>   
>   	/*
> @@ -154,7 +161,7 @@ void kvm_xen_update_runstate_guest(struct kvm_vcpu *v, int state)
>   					  &vx->current_runstate,
>   					  offsetof(struct vcpu_runstate_info, state),
>   					  sizeof(vx->current_runstate)))
> -		return;
> +		goto out;
>   
>   	/*
>   	 * Write the actual runstate times immediately after the
> @@ -173,7 +180,7 @@ void kvm_xen_update_runstate_guest(struct kvm_vcpu *v, int state)
>   					  &vx->runstate_times[0],
>   					  offset + sizeof(u64),
>   					  sizeof(vx->runstate_times)))
> -		return;
> +		goto out;
>   
>   	smp_wmb();
>   
> @@ -186,7 +193,10 @@ void kvm_xen_update_runstate_guest(struct kvm_vcpu *v, int state)
>   	if (kvm_write_guest_offset_cached(v->kvm, &v->arch.xen.runstate_cache,
>   					  &state_entry_time, offset,
>   					  sizeof(state_entry_time)))
> -		return;
> +		goto out;
> +
> +out:
> +	srcu_read_unlock(&v->kvm->srcu, idx);
>   }
>   
>   int __kvm_xen_has_interrupt(struct kvm_vcpu *v)
>