Re: [PATCH 2/3] kernel,cred,kvm,security - removing superfluous rcu_read_lock_held check

From: Paolo Bonzini <pbonzini@redhat.com>
To: Jiri Olsa <jolsa@redhat.com>
Cc: dhowells@redhat.com, lizf@cn.fujitsu.com, paul.moore@hp.com,
	johannes@sipsolutions.net, dipankar@in.ibm.com,
	linux-kernel@vger.kernel.org,
	containers@lists.linux-foundation.org, kvm@vger.kernel.org
Subject: Re: [PATCH 2/3] kernel,cred,kvm,security - removing superfluous rcu_read_lock_held check
Date: Mon, 01 Nov 2010 23:42:23 +0100	[thread overview]
Message-ID: <4CCF424F.3050205@redhat.com> (raw)
In-Reply-To: <1288638909-10541-3-git-send-email-jolsa@redhat.com>

On 11/01/2010 08:15 PM, Jiri Olsa wrote:
> diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
> index a055742..a90a7e3 100644
> --- a/include/linux/kvm_host.h
> +++ b/include/linux/kvm_host.h
> @@ -256,7 +256,6 @@ void kvm_put_kvm(struct kvm *kvm);
>   static inline struct kvm_memslots *kvm_memslots(struct kvm *kvm)
>   {
>   	return rcu_dereference_check(kvm->memslots,
> -			srcu_read_lock_held(&kvm->srcu)
>   			|| lockdep_is_held(&kvm->slots_lock));
>   }
>

This is an srcu_read_lock_held, which you don't touch here:

> diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c
> index 9d8e8fb..0956a73 100644
> --- a/kernel/rcutorture.c
> +++ b/kernel/rcutorture.c
> @@ -807,7 +807,6 @@ static void rcu_torture_timer(unsigned long unused)
>  	idx = cur_ops->readlock();
>  	completed = cur_ops->completed();
>  	p = rcu_dereference_check(rcu_torture_current,
> -				  rcu_read_lock_held() ||
>  				  rcu_read_lock_bh_held() ||
>  				  rcu_read_lock_sched_held() ||
>  				  srcu_read_lock_held(&srcu_ctl));

I guess the kvm hunk is the incorrect one?

Paolo