Re: Real-Time Preemption and RCU

From: Ingo Molnar <mingo@elte.hu>
To: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: dipankar@in.ibm.com, shemminger@osdl.org, akpm@osdl.org,
	torvalds@osdl.org, rusty@au1.ibm.com, tgall@us.ibm.com,
	jim.houston@comcast.net, manfred@colorfullife.com, gh@us.ibm.com,
	linux-kernel@vger.kernel.org
Subject: Re: Real-Time Preemption and RCU
Date: Fri, 18 Mar 2005 11:03:39 +0100	[thread overview]
Message-ID: <20050318100339.GA15386@elte.hu> (raw)
In-Reply-To: <20050318002026.GA2693@us.ibm.com>

there's a problem in #5's rcu_read_lock():

        void
        rcu_read_lock(void)
        {
                preempt_disable();
                if (current->rcu_read_lock_nesting++ == 0) {
                        current->rcu_read_lock_ptr =
                                &__get_cpu_var(rcu_data).lock;
                        read_lock(current->rcu_read_lock_ptr);
                }
                preempt_enable();
        }

not only are read_lock()-ed sections preemptible, read_lock() itself may
block, so it cannot be called from within preempt_disable(). How about
something like:

        void
        rcu_read_lock(void)
        {
                preempt_disable();
                if (current->rcu_read_lock_nesting++ == 0) {
                        current->rcu_read_lock_ptr =
                                &__get_cpu_var(rcu_data).lock;
                        preempt_enable();
                        read_lock(current->rcu_read_lock_ptr);
                } else
                        preempt_enable();
        }

this would still make it 'statistically scalable' - but is it correct?

	Ingo