信号量有一个很明显的缺点，没有区分临界区的读写属性，读写锁允许多个线程进程并发的访问临界区，但是写访问只限于一个线程，在多处理器系统中允许多个读者访问共享资源，但是写者有排他性，读写锁的特性如下：允许多个读者同时访问临界区，但是同一时间不能进入；同一时刻只允许一个写者进入临界区；读者和写者不能同时进入临界区。读写锁也有关闭中断和下半部的版本。

RCU:read-copy-update   。。。。。。。。。。。。。。。。。。。。

问题：rcu相比读写锁，解决了什么问题？ rcu的基本原理？

1、由于内核中spinlock mutex 等都使用了原子操作指令，即原子的访问内存，但是当多cpu 竞争访问临界区时会让cpu的cache命中率下降，性能下降。同时读写锁有个缺陷，读者和写者不能同时存在。

rcu实现的目标就是要解决这个问题，为了使线程同步开销小。不需要原子操作以及内存屏障而访问数据，把同步的问题交给写者线程，写者线程等待所有的读者线程完成后才会吧旧数据销毁。当有多个写者线程存在时，需要额外的保护机制。

RCU原理：简单理解为 记录了所有指向共享数据的指针使用者，当要修改共享数据时，先创建一个副本，在副本中修改。所有读者离开临界区后，指针指向新的修改副本后的地方，并且删除旧数据。

目前在内核中链表使用RCU较多。

在经典RCU中，RCU读侧临界部分由rcu_read_lock() 和rcu_read_unlock()界定，它们可以嵌套。

对应的同步更新原语为synchronize_rcu()，还有同义的synchronize_net()，等待当前正执行的RCU读侧闻临界部分运行完成。等待的时间称为“宽限期”。

异步更新侧原语call_rcu()在宽限期之后触发指定的函数，如：call_rcu(p,f)调用触发回调函数f(p)。有些情况，如：当卸载使用call_rcu()的模块，必须等待所有RCU回调函数完成，原语rcu_barrier()起该作用。
在“RCU BH”列中，rcu_read_lock_bh() 和rcu_read_unlock_bh()界定读侧临界部分，call_rcu_bh()在宽限期后触发指定的函数。注意：RCU BH没有同步接口synchronize_rcu_bh()，如果需要，用户很容易添加同步接口函数。

直接操作指针的原语rcu_assign_pointer()和rcu_dereference()用于创建RCU保护的非链表数据结构，如：数组和树

/*
Please note that the "What is RCU?" LWN series is an excellent place
to start learning about RCU:

1. What is RCU, Fundamentally? http://lwn.net/Articles/262464/
2. What is RCU? Part 2: Usage http://lwn.net/Articles/263130/
3. RCU part 3: the RCU API http://lwn.net/Articles/264090/
4. The RCU API, 2010 Edition http://lwn.net/Articles/418853/

What is RCU?

RCU is a synchronization mechanism that was added to the Linux kernel
during the 2.5 development effort that is optimized for read-mostly
situations. Although RCU is actually quite simple once you understand it,
getting there can sometimes be a challenge. Part of the problem is that
most of the past descriptions of RCU have been written with the mistaken
assumption that there is "one true way" to describe RCU. Instead,
the experience has been that different people must take different paths
to arrive at an understanding of RCU. This document provides several
different paths, as follows:

1. RCU OVERVIEW
2. WHAT IS RCU'S CORE API?
3. WHAT ARE SOME EXAMPLE USES OF CORE RCU API?
4. WHAT IF MY UPDATING THREAD CANNOT BLOCK?
5. WHAT ARE SOME SIMPLE IMPLEMENTATIONS OF RCU?
6. ANALOGY WITH READER-WRITER LOCKING
7. FULL LIST OF RCU APIs
8. ANSWERS TO QUICK QUIZZES

People who prefer starting with a conceptual overview should focus on
Section 1, though most readers will profit by reading this section at
some point. People who prefer to start with an API that they can then
experiment with should focus on Section 2. People who prefer to start
with example uses should focus on Sections 3 and 4. People who need to
understand the RCU implementation should focus on Section 5, then dive
into the kernel source code. People who reason best by analogy should
focus on Section 6. Section 7 serves as an index to the docbook API
documentation, and Section 8 is the traditional answer key.

So, start with the section that makes the most sense to you and your
preferred method of learning. If you need to know everything about
everything, feel free to read the whole thing -- but if you are really
that type of person, you have perused the source code and will therefore
never need this document anyway. ;-)

1. RCU OVERVIEW

The basic idea behind RCU is to split updates into "removal" and
"reclamation" phases. The removal phase removes references to data items
within a data structure (possibly by replacing them with references to
new versions of these data items), and can run concurrently with readers.
The reason that it is safe to run the removal phase concurrently with
readers is the semantics of modern CPUs guarantee that readers will see
either the old or the new version of the data structure rather than a
partially updated reference. The reclamation phase does the work of reclaiming
(e.g., freeing) the data items removed from the data structure during the
removal phase. Because reclaiming data items can disrupt any readers
concurrently referencing those data items, the reclamation phase must
not start until readers no longer hold references to those data items.

Splitting the update into removal and reclamation phases permits the
updater to perform the removal phase immediately, and to defer the
reclamation phase until all readers active during the removal phase have
completed, either by blocking until they finish or by registering a
callback that is invoked after they finish. Only readers that are active
during the removal phase need be considered, because any reader starting
after the removal phase will be unable to gain a reference to the removed
data items, and therefore cannot be disrupted by the reclamation phase.

So the typical RCU update sequence goes something like the following:

a. Remove pointers to a data structure, so that subsequent
readers cannot gain a reference to it.

b. Wait for all previous readers to complete their RCU read-side
critical sections.

c. At this point, there cannot be any readers who hold references
to the data structure, so it now may safely be reclaimed
(e.g., kfree()d).

Step (b) above is the key idea underlying RCU's deferred destruction.
The ability to wait until all readers are done allows RCU readers to
use much lighter-weight synchronization, in some cases, absolutely no
synchronization at all. In contrast, in more conventional lock-based
schemes, readers must use heavy-weight synchronization in order to
prevent an updater from deleting the data structure out from under them.
This is because lock-based updaters typically update data items in place,
and must therefore exclude readers. In contrast, RCU-based updaters
typically take advantage of the fact that writes to single aligned
pointers are atomic on modern CPUs, allowing atomic insertion, removal,
and replacement of data items in a linked structure without disrupting
readers. Concurrent RCU readers can then continue accessing the old
versions, and can dispense with the atomic operations, memory barriers,
and communications cache misses that are so expensive on present-day
SMP computer systems, even in absence of lock contention.

In the three-step procedure shown above, the updater is performing both
the removal and the reclamation step, but it is often helpful for an
entirely different thread to do the reclamation, as is in fact the case
in the Linux kernel's directory-entry cache (dcache). Even if the same
thread performs both the update step (step (a) above) and the reclamation
step (step (c) above), it is often helpful to think of them separately.
For example, RCU readers and updaters need not communicate at all,
but RCU provides implicit low-overhead communication between readers
and reclaimers, namely, in step (b) above.

So how the heck can a reclaimer tell when a reader is done, given
that readers are not doing any sort of synchronization operations???
Read on to learn about how RCU's API makes this easy.

2. WHAT IS RCU'S CORE API?

The core RCU API is quite small:

a. rcu_read_lock()
b. rcu_read_unlock()
c. synchronize_rcu() / call_rcu()
d. rcu_assign_pointer()
e. rcu_dereference()

There are many other members of the RCU API, but the rest can be
expressed in terms of these five, though most implementations instead
express synchronize_rcu() in terms of the call_rcu() callback API.

The five core RCU APIs are described below, the other 18 will be enumerated
later. See the kernel docbook documentation for more info, or look directly
at the function header comments.

rcu_read_lock()

void rcu\_read\_lock(void);

Used by a reader to inform the reclaimer that the reader is  
entering an RCU read-side critical section.  It is illegal  
to block while in an RCU read-side critical section, though  
kernels built with CONFIG\_PREEMPT\_RCU can preempt RCU  
read-side critical sections.  Any RCU-protected data structure  
accessed during an RCU read-side critical section is guaranteed to  
remain unreclaimed for the full duration of that critical section.  
Reference counts may be used in conjunction with RCU to maintain  
longer-term references to data structures.

rcu_read_unlock()

void rcu\_read\_unlock(void);

Used by a reader to inform the reclaimer that the reader is  
exiting an RCU read-side critical section.  Note that RCU  
read-side critical sections may be nested and/or overlapping.

synchronize_rcu()

void synchronize\_rcu(void);

Marks the end of updater code and the beginning of reclaimer  
code.  It does this by blocking until all pre-existing RCU  
read-side critical sections on all CPUs have completed.  
Note that synchronize\_rcu() will -not- necessarily wait for  
any subsequent RCU read-side critical sections to complete.  
For example, consider the following sequence of events:

         CPU 0                  CPU 1                 CPU 2  
     ----------------- ------------------------- ---------------  
 1.  rcu\_read\_lock()  
 2.                    enters synchronize\_rcu()  
 3.                                               rcu\_read\_lock()  
 4.  rcu\_read\_unlock()  
 5.                     exits synchronize\_rcu()  
 6.                                              rcu\_read\_unlock()

To reiterate, synchronize\_rcu() waits only for ongoing RCU  
read-side critical sections to complete, not necessarily for  
any that begin after synchronize\_rcu() is invoked.

Of course, synchronize\_rcu() does not necessarily return  
-immediately- after the last pre-existing RCU read-side critical  
section completes.  For one thing, there might well be scheduling  
delays.  For another thing, many RCU implementations process  
requests in batches in order to improve efficiencies, which can  
further delay synchronize\_rcu().

Since synchronize\_rcu() is the API that must figure out when  
readers are done, its implementation is key to RCU.  For RCU  
to be useful in all but the most read-intensive situations,  
synchronize\_rcu()'s overhead must also be quite small.

The call\_rcu() API is a callback form of synchronize\_rcu(),  
and is described in more detail in a later section.  Instead of  
blocking, it registers a function and argument which are invoked  
after all ongoing RCU read-side critical sections have completed.  
This callback variant is particularly useful in situations where  
it is illegal to block or where update-side performance is  
critically important.

However, the call\_rcu() API should not be used lightly, as use  
of the synchronize\_rcu() API generally results in simpler code.  
In addition, the synchronize\_rcu() API has the nice property  
of automatically limiting update rate should grace periods  
be delayed.  This property results in system resilience in face  
of denial-of-service attacks.  Code using call\_rcu() should limit  
update rate in order to gain this same sort of resilience.  See  
checklist.txt for some approaches to limiting the update rate.

rcu_assign_pointer()

typeof(p) rcu\_assign\_pointer(p, typeof(p) v);

Yes, rcu\_assign\_pointer() -is- implemented as a macro, though it  
would be cool to be able to declare a function in this manner.  
(Compiler experts will no doubt disagree.)

The updater uses this function to assign a new value to an  
RCU-protected pointer, in order to safely communicate the change  
in value from the updater to the reader.  This function returns  
the new value, and also executes any memory-barrier instructions  
required for a given CPU architecture.

Perhaps just as important, it serves to document (1) which  
pointers are protected by RCU and (2) the point at which a  
given structure becomes accessible to other CPUs.  That said,  
rcu\_assign\_pointer() is most frequently used indirectly, via  
the \_rcu list-manipulation primitives such as list\_add\_rcu().

rcu_dereference()

typeof(p) rcu\_dereference(p);

Like rcu\_assign\_pointer(), rcu\_dereference() must be implemented  
as a macro.

The reader uses rcu\_dereference() to fetch an RCU-protected  
pointer, which returns a value that may then be safely  
dereferenced.  Note that rcu\_deference() does not actually  
dereference the pointer, instead, it protects the pointer for  
later dereferencing.  It also executes any needed memory-barrier  
instructions for a given CPU architecture.  Currently, only Alpha  
needs memory barriers within rcu\_dereference() -- on other CPUs,  
it compiles to nothing, not even a compiler directive.

Common coding practice uses rcu\_dereference() to copy an  
RCU-protected pointer to a local variable, then dereferences  
this local variable, for example as follows:

    p = rcu\_dereference(head.next);  
    return p->data;

However, in this case, one could just as easily combine these  
into one statement:

    return rcu\_dereference(head.next)->data;

If you are going to be fetching multiple fields from the  
RCU-protected structure, using the local variable is of  
course preferred.  Repeated rcu\_dereference() calls look  
ugly, do not guarantee that the same pointer will be returned  
if an update happened while in the critical section, and incur  
unnecessary overhead on Alpha CPUs.

Note that the value returned by rcu\_dereference() is valid  
only within the enclosing RCU read-side critical section.  
For example, the following is -not- legal:

    rcu\_read\_lock();  
    p = rcu\_dereference(head.next);  
    rcu\_read\_unlock();  
    x = p->address;    /\* BUG!!! \*/  
    rcu\_read\_lock();  
    y = p->data;    /\* BUG!!! \*/  
    rcu\_read\_unlock();

Holding a reference from one RCU read-side critical section  
to another is just as illegal as holding a reference from  
one lock-based critical section to another!  Similarly,  
using a reference outside of the critical section in which  
it was acquired is just as illegal as doing so with normal  
locking.

As with rcu\_assign\_pointer(), an important function of  
rcu\_dereference() is to document which pointers are protected by  
RCU, in particular, flagging a pointer that is subject to changing  
at any time, including immediately after the rcu\_dereference().  
And, again like rcu\_assign\_pointer(), rcu\_dereference() is  
typically used indirectly, via the \_rcu list-manipulation  
primitives, such as list\_for\_each\_entry\_rcu().

The following diagram shows how each API communicates among the
reader, updater, and reclaimer.

    rcu\_assign\_pointer()  
                    +--------+  
    +---------------------->| reader |---------+  
    |                       +--------+         |  
    |                           |              |  
    |                           |              | Protect:  
    |                           |              | rcu\_read\_lock()  
    |                           |              | rcu\_read\_unlock()  
    |        rcu\_dereference()  |              |  
   +---------+                      |              |  
   | updater |<---------------------+              |  
   +---------+                                     V  
    |                                    +-----------+  
    +----------------------------------->| reclaimer |  
                             +-----------+  
      Defer:  
      synchronize\_rcu() & call\_rcu()

The RCU infrastructure observes the time sequence of rcu_read_lock(),
rcu_read_unlock(), synchronize_rcu(), and call_rcu() invocations in
order to determine when (1) synchronize_rcu() invocations may return
to their callers and (2) call_rcu() callbacks may be invoked. Efficient
implementations of the RCU infrastructure make heavy use of batching in
order to amortize their overhead over many uses of the corresponding APIs.

There are no fewer than three RCU mechanisms in the Linux kernel; the
diagram above shows the first one, which is by far the most commonly used.
The rcu_dereference() and rcu_assign_pointer() primitives are used for
all three mechanisms, but different defer and protect primitives are
used as follows:

Defer            Protect

a. synchronize_rcu() rcu_read_lock() / rcu_read_unlock()
call_rcu() rcu_dereference()

b. synchronize_rcu_bh() rcu_read_lock_bh() / rcu_read_unlock_bh()
call_rcu_bh() rcu_dereference_bh()

c. synchronize_sched() rcu_read_lock_sched() / rcu_read_unlock_sched()
call_rcu_sched() preempt_disable() / preempt_enable()
local_irq_save() / local_irq_restore()
hardirq enter / hardirq exit
NMI enter / NMI exit
rcu_dereference_sched()

These three mechanisms are used as follows:

a. RCU applied to normal data structures.

b. RCU applied to networking data structures that may be subjected
to remote denial-of-service attacks.

c. RCU applied to scheduler and interrupt/NMI-handler tasks.

Again, most uses will be of (a). The (b) and (c) cases are important
for specialized uses, but are relatively uncommon.

3. WHAT ARE SOME EXAMPLE USES OF CORE RCU API?

This section shows a simple use of the core RCU API to protect a
global pointer to a dynamically allocated structure. More-typical
uses of RCU may be found in listRCU.txt, arrayRCU.txt, and NMI-RCU.txt.

struct foo {  
    int a;  
    char b;  
    long c;  
};  
DEFINE\_SPINLOCK(foo\_mutex);

struct foo \_\_rcu \*gbl\_foo;

/\*  
 \* Create a new struct foo that is the same as the one currently  
 \* pointed to by gbl\_foo, except that field "a" is replaced  
 \* with "new\_a".  Points gbl\_foo to the new structure, and  
 \* frees up the old structure after a grace period.  
 \*  
 \* Uses rcu\_assign\_pointer() to ensure that concurrent readers  
 \* see the initialized version of the new structure.  
 \*  
 \* Uses synchronize\_rcu() to ensure that any readers that might  
 \* have references to the old structure complete before freeing  
 \* the old structure.  
 \*/  
void foo\_update\_a(int new\_a)  
{  
    struct foo \*new\_fp;  
    struct foo \*old\_fp;

    new\_fp = kmalloc(sizeof(\*new\_fp), GFP\_KERNEL);  
    spin\_lock(&foo\_mutex);  
    old\_fp = rcu\_dereference\_protected(gbl\_foo, lockdep\_is\_held(&foo\_mutex));  
    \*new\_fp = \*old\_fp;  
    new\_fp->a = new\_a;  
    rcu\_assign\_pointer(gbl\_foo, new\_fp);  
    spin\_unlock(&foo\_mutex);  
    synchronize\_rcu();  
    kfree(old\_fp);  
}

/\*  
 \* Return the value of field "a" of the current gbl\_foo  
 \* structure.  Use rcu\_read\_lock() and rcu\_read\_unlock()  
 \* to ensure that the structure does not get deleted out  
 \* from under us, and use rcu\_dereference() to ensure that  
 \* we see the initialized version of the structure (important  
 \* for DEC Alpha and for people reading the code).  
 \*/  
int foo\_get\_a(void)  
{  
    int retval;

    rcu\_read\_lock();  
    retval = rcu\_dereference(gbl\_foo)->a;  
    rcu\_read\_unlock();  
    return retval;  
}

So, to sum up:

o Use rcu_read_lock() and rcu_read_unlock() to guard RCU
read-side critical sections.

o Within an RCU read-side critical section, use rcu_dereference()
to dereference RCU-protected pointers.

o Use some solid scheme (such as locks or semaphores) to
keep concurrent updates from interfering with each other.

o Use rcu_assign_pointer() to update an RCU-protected pointer.
This primitive protects concurrent readers from the updater,
-not- concurrent updates from each other! You therefore still
need to use locking (or something similar) to keep concurrent
rcu_assign_pointer() primitives from interfering with each other.

o Use synchronize_rcu() -after- removing a data element from an
RCU-protected data structure, but -before- reclaiming/freeing
the data element, in order to wait for the completion of all
RCU read-side critical sections that might be referencing that
data item.

See checklist.txt for additional rules to follow when using RCU.
And again, more-typical uses of RCU may be found in listRCU.txt,
arrayRCU.txt, and NMI-RCU.txt.

4. WHAT IF MY UPDATING THREAD CANNOT BLOCK?

In the example above, foo_update_a() blocks until a grace period elapses.
This is quite simple, but in some cases one cannot afford to wait so
long -- there might be other high-priority work to be done.

In such cases, one uses call_rcu() rather than synchronize_rcu().
The call_rcu() API is as follows:

void call\_rcu(struct rcu\_head \* head,  
          void (\*func)(struct rcu\_head \*head));

This function invokes func(head) after a grace period has elapsed.
This invocation might happen from either softirq or process context,
so the function is not permitted to block. The foo struct needs to
have an rcu_head structure added, perhaps as follows:

struct foo {  
    int a;  
    char b;  
    long c;  
    struct rcu\_head rcu;  
};

The foo_update_a() function might then be written as follows:

/\*  
 \* Create a new struct foo that is the same as the one currently  
 \* pointed to by gbl\_foo, except that field "a" is replaced  
 \* with "new\_a".  Points gbl\_foo to the new structure, and  
 \* frees up the old structure after a grace period.  
 \*  
 \* Uses rcu\_assign\_pointer() to ensure that concurrent readers  
 \* see the initialized version of the new structure.  
 \*  
 \* Uses call\_rcu() to ensure that any readers that might have  
 \* references to the old structure complete before freeing the  
 \* old structure.  
 \*/  
void foo\_update\_a(int new\_a)  
{  
    struct foo \*new\_fp;  
    struct foo \*old\_fp;

    new\_fp = kmalloc(sizeof(\*new\_fp), GFP\_KERNEL);  
    spin\_lock(&foo\_mutex);  
    old\_fp = rcu\_dereference\_protected(gbl\_foo, lockdep\_is\_held(&foo\_mutex));  
    \*new\_fp = \*old\_fp;  
    new\_fp->a = new\_a;  
    rcu\_assign\_pointer(gbl\_foo, new\_fp);  
    spin\_unlock(&foo\_mutex);  
    call\_rcu(&old\_fp->rcu, foo\_reclaim);  
}

The foo_reclaim() function might appear as follows:

void foo\_reclaim(struct rcu\_head \*rp)  
{  
    struct foo \*fp = container\_of(rp, struct foo, rcu);

    foo\_cleanup(fp->a);

    kfree(fp);  
}

The container_of() primitive is a macro that, given a pointer into a
struct, the type of the struct, and the pointed-to field within the
struct, returns a pointer to the beginning of the struct.

The use of call_rcu() permits the caller of foo_update_a() to
immediately regain control, without needing to worry further about the
old version of the newly updated element. It also clearly shows the
RCU distinction between updater, namely foo_update_a(), and reclaimer,
namely foo_reclaim().

The summary of advice is the same as for the previous section, except
that we are now using call_rcu() rather than synchronize_rcu():

o Use call_rcu() -after- removing a data element from an
RCU-protected data structure in order to register a callback
function that will be invoked after the completion of all RCU
read-side critical sections that might be referencing that
data item.

If the callback for call_rcu() is not doing anything more than calling
kfree() on the structure, you can use kfree_rcu() instead of call_rcu()
to avoid having to write your own callback:

kfree\_rcu(old\_fp, rcu);

Again, see checklist.txt for additional rules governing the use of RCU.

5. WHAT ARE SOME SIMPLE IMPLEMENTATIONS OF RCU?

One of the nice things about RCU is that it has extremely simple "toy"
implementations that are a good first step towards understanding the
production-quality implementations in the Linux kernel. This section
presents two such "toy" implementations of RCU, one that is implemented
in terms of familiar locking primitives, and another that more closely
resembles "classic" RCU. Both are way too simple for real-world use,
lacking both functionality and performance. However, they are useful
in getting a feel for how RCU works. See kernel/rcupdate.c for a
production-quality implementation, and see:

http://www.rdrop.com/users/paulmck/RCU

for papers describing the Linux kernel RCU implementation. The OLS'01
and OLS'02 papers are a good introduction, and the dissertation provides
more details on the current implementation as of early 2004.

5A. "TOY" IMPLEMENTATION #1: LOCKING

This section presents a "toy" RCU implementation that is based on
familiar locking primitives. Its overhead makes it a non-starter for
real-life use, as does its lack of scalability. It is also unsuitable
for realtime use, since it allows scheduling latency to "bleed" from
one read-side critical section to another.

However, it is probably the easiest implementation to relate to, so is
a good starting point.

It is extremely simple:

static DEFINE\_RWLOCK(rcu\_gp\_mutex);

void rcu\_read\_lock(void)  
{  
    read\_lock(&rcu\_gp\_mutex);  
}

void rcu\_read\_unlock(void)  
{  
    read\_unlock(&rcu\_gp\_mutex);  
}

void synchronize\_rcu(void)  
{  
    write\_lock(&rcu\_gp\_mutex);  
    write\_unlock(&rcu\_gp\_mutex);  
}

[You can ignore rcu_assign_pointer() and rcu_dereference() without
missing much. But here they are anyway. And whatever you do, don't
forget about them when submitting patches making use of RCU!]

#define rcu\_assign\_pointer(p, v)    ({ \\  
                        smp\_wmb(); \\  
                        (p) = (v); \\  
                    })

#define rcu\_dereference(p)     ({ \\  
                typeof(p) \_\_\_\_\_\_\_\_\_p1 = p; \\  
                smp\_read\_barrier\_depends(); \\  
                (\_\_\_\_\_\_\_\_\_p1); \\  
                })

The rcu_read_lock() and rcu_read_unlock() primitive read-acquire
and release a global reader-writer lock. The synchronize_rcu()
primitive write-acquires this same lock, then immediately releases
it. This means that once synchronize_rcu() exits, all RCU read-side
critical sections that were in progress before synchronize_rcu() was
called are guaranteed to have completed -- there is no way that
synchronize_rcu() would have been able to write-acquire the lock
otherwise.

It is possible to nest rcu_read_lock(), since reader-writer locks may
be recursively acquired. Note also that rcu_read_lock() is immune
from deadlock (an important property of RCU). The reason for this is
that the only thing that can block rcu_read_lock() is a synchronize_rcu().
But synchronize_rcu() does not acquire any locks while holding rcu_gp_mutex,
so there can be no deadlock cycle.

Quick Quiz #1: Why is this argument naive? How could a deadlock
occur when using this algorithm in a real-world Linux
kernel? How could this deadlock be avoided?

5B. "TOY" EXAMPLE #2: CLASSIC RCU

This section presents a "toy" RCU implementation that is based on
"classic RCU". It is also short on performance (but only for updates) and
on features such as hotplug CPU and the ability to run in CONFIG_PREEMPT
kernels. The definitions of rcu_dereference() and rcu_assign_pointer()
are the same as those shown in the preceding section, so they are omitted.

void rcu\_read\_lock(void) { }

void rcu\_read\_unlock(void) { }

void synchronize\_rcu(void)  
{  
    int cpu;

    for\_each\_possible\_cpu(cpu)  
        run\_on(cpu);  
}

Note that rcu_read_lock() and rcu_read_unlock() do absolutely nothing.
This is the great strength of classic RCU in a non-preemptive kernel:
read-side overhead is precisely zero, at least on non-Alpha CPUs.
And there is absolutely no way that rcu_read_lock() can possibly
participate in a deadlock cycle!

The implementation of synchronize_rcu() simply schedules itself on each
CPU in turn. The run_on() primitive can be implemented straightforwardly
in terms of the sched_setaffinity() primitive. Of course, a somewhat less
"toy" implementation would restore the affinity upon completion rather
than just leaving all tasks running on the last CPU, but when I said
"toy", I meant -toy-!

So how the heck is this supposed to work???

Remember that it is illegal to block while in an RCU read-side critical
section. Therefore, if a given CPU executes a context switch, we know
that it must have completed all preceding RCU read-side critical sections.
Once -all- CPUs have executed a context switch, then -all- preceding
RCU read-side critical sections will have completed.

So, suppose that we remove a data item from its structure and then invoke
synchronize_rcu(). Once synchronize_rcu() returns, we are guaranteed
that there are no RCU read-side critical sections holding a reference
to that data item, so we can safely reclaim it.

Quick Quiz #2: Give an example where Classic RCU's read-side
overhead is -negative-.

Quick Quiz #3: If it is illegal to block in an RCU read-side
critical section, what the heck do you do in
PREEMPT_RT, where normal spinlocks can block???

6. ANALOGY WITH READER-WRITER LOCKING

Although RCU can be used in many different ways, a very common use of
RCU is analogous to reader-writer locking. The following unified
diff shows how closely related RCU and reader-writer locking can be.

@@ -13,15 +14,15 @@  
    struct list\_head \*lp;  
    struct el \*p;

-    read\_lock();  
-    list\_for\_each\_entry(p, head, lp) {  
+    rcu\_read\_lock();  
+    list\_for\_each\_entry\_rcu(p, head, lp) {  
        if (p->key == key) {  
            \*result = p->data;  
-            read\_unlock();  
+            rcu\_read\_unlock();  
            return 1;  
        }  
    }  
-    read\_unlock();  
+    rcu\_read\_unlock();  
    return 0;  
 }

@@ -29,15 +30,16 @@  
 {  
    struct el \*p;

-    write\_lock(&listmutex);  
+    spin\_lock(&listmutex);  
    list\_for\_each\_entry(p, head, lp) {  
        if (p->key == key) {  
-            list\_del(&p->list);  
-            write\_unlock(&listmutex);  
+            list\_del\_rcu(&p->list);  
+            spin\_unlock(&listmutex);  
+            synchronize\_rcu();  
            kfree(p);  
            return 1;  
        }  
    }  
-    write\_unlock(&listmutex);  
+    spin\_unlock(&listmutex);  
    return 0;  
 }

Or, for those who prefer a side-by-side listing:

1 struct el { 1 struct el {
2 struct list_head list; 2 struct list_head list;
3 long key; 3 long key;
4 spinlock_t mutex; 4 spinlock_t mutex;
5 int data; 5 int data;
6 /* Other data fields */ 6 /* Other data fields */
7 }; 7 };
8 spinlock_t listmutex; 8 spinlock_t listmutex;
9 struct el head; 9 struct el head;

1 int search(long key, int *result) 1 int search(long key, int *result)
2 { 2 {
3 struct list_head *lp; 3 struct list_head *lp;
4 struct el *p; 4 struct el *p;
5 5
6 read_lock(); 6 rcu_read_lock();
7 list_for_each_entry(p, head, lp) { 7 list_for_each_entry_rcu(p, head, lp) {
8 if (p->key == key) { 8 if (p->key == key) {
9 *result = p->data; 9 *result = p->data;
10 read_unlock(); 10 rcu_read_unlock();
11 return 1; 11 return 1;
12 } 12 }
13 } 13 }
14 read_unlock(); 14 rcu_read_unlock();
15 return 0; 15 return 0;
16 } 16 }

1 int delete(long key) 1 int delete(long key)
2 { 2 {
3 struct el *p; 3 struct el *p;
4 4
5 write_lock(&listmutex); 5 spin_lock(&listmutex);
6 list_for_each_entry(p, head, lp) { 6 list_for_each_entry(p, head, lp) {
7 if (p->key == key) { 7 if (p->key == key) {
8 list_del(&p->list); 8 list_del_rcu(&p->list);
9 write_unlock(&listmutex); 9 spin_unlock(&listmutex);
10 synchronize_rcu();
10 kfree(p); 11 kfree(p);
11 return 1; 12 return 1;
12 } 13 }
13 } 14 }
14 write_unlock(&listmutex); 15 spin_unlock(&listmutex);
15 return 0; 16 return 0;
16 } 17 }

Either way, the differences are quite small. Read-side locking moves
to rcu_read_lock() and rcu_read_unlock, update-side locking moves from
a reader-writer lock to a simple spinlock, and a synchronize_rcu()
precedes the kfree().

However, there is one potential catch: the read-side and update-side
critical sections can now run concurrently. In many cases, this will
not be a problem, but it is necessary to check carefully regardless.
For example, if multiple independent list updates must be seen as
a single atomic update, converting to RCU will require special care.

Also, the presence of synchronize_rcu() means that the RCU version of
delete() can now block. If this is a problem, there is a callback-based
mechanism that never blocks, namely call_rcu() or kfree_rcu(), that can
be used in place of synchronize_rcu().

7. FULL LIST OF RCU APIs

The RCU APIs are documented in docbook-format header comments in the
Linux-kernel source code, but it helps to have a full list of the
APIs, since there does not appear to be a way to categorize them
in docbook. Here is the list, by category.

RCU list traversal:

list\_entry\_rcu  
list\_first\_entry\_rcu  
list\_next\_rcu  
list\_for\_each\_entry\_rcu  
list\_for\_each\_entry\_continue\_rcu  
hlist\_first\_rcu  
hlist\_next\_rcu  
hlist\_pprev\_rcu  
hlist\_for\_each\_entry\_rcu  
hlist\_for\_each\_entry\_rcu\_bh  
hlist\_for\_each\_entry\_continue\_rcu  
hlist\_for\_each\_entry\_continue\_rcu\_bh  
hlist\_nulls\_first\_rcu  
hlist\_nulls\_for\_each\_entry\_rcu  
hlist\_bl\_first\_rcu  
hlist\_bl\_for\_each\_entry\_rcu

RCU pointer/list update:

rcu\_assign\_pointer  
list\_add\_rcu  
list\_add\_tail\_rcu  
list\_del\_rcu  
list\_replace\_rcu  
hlist\_add\_behind\_rcu  
hlist\_add\_before\_rcu  
hlist\_add\_head\_rcu  
hlist\_del\_rcu  
hlist\_del\_init\_rcu  
hlist\_replace\_rcu  
list\_splice\_init\_rcu()  
hlist\_nulls\_del\_init\_rcu  
hlist\_nulls\_del\_rcu  
hlist\_nulls\_add\_head\_rcu  
hlist\_bl\_add\_head\_rcu  
hlist\_bl\_del\_init\_rcu  
hlist\_bl\_del\_rcu  
hlist\_bl\_set\_first\_rcu

RCU: Critical sections Grace period Barrier

rcu\_read\_lock        synchronize\_net        rcu\_barrier  
rcu\_read\_unlock        synchronize\_rcu  
rcu\_dereference        synchronize\_rcu\_expedited  
rcu\_read\_lock\_held    call\_rcu  
rcu\_dereference\_check    kfree\_rcu  
rcu\_dereference\_protected

bh: Critical sections Grace period Barrier

rcu\_read\_lock\_bh    call\_rcu\_bh        rcu\_barrier\_bh  
rcu\_read\_unlock\_bh    synchronize\_rcu\_bh  
rcu\_dereference\_bh    synchronize\_rcu\_bh\_expedited  
rcu\_dereference\_bh\_check  
rcu\_dereference\_bh\_protected  
rcu\_read\_lock\_bh\_held

sched: Critical sections Grace period Barrier

rcu\_read\_lock\_sched    synchronize\_sched    rcu\_barrier\_sched  
rcu\_read\_unlock\_sched    call\_rcu\_sched  
\[preempt\_disable\]    synchronize\_sched\_expedited  
\[and friends\]  
rcu\_read\_lock\_sched\_notrace  
rcu\_read\_unlock\_sched\_notrace  
rcu\_dereference\_sched  
rcu\_dereference\_sched\_check  
rcu\_dereference\_sched\_protected  
rcu\_read\_lock\_sched\_held

SRCU: Critical sections Grace period Barrier

srcu\_read\_lock        synchronize\_srcu    srcu\_barrier  
srcu\_read\_unlock    call\_srcu  
srcu\_dereference    synchronize\_srcu\_expedited  
srcu\_dereference\_check  
srcu\_read\_lock\_held

SRCU: Initialization/cleanup
init_srcu_struct
cleanup_srcu_struct

All: lockdep-checked RCU-protected pointer access

rcu\_access\_pointer  
rcu\_dereference\_raw  
RCU\_LOCKDEP\_WARN  
rcu\_sleep\_check  
RCU\_NONIDLE

See the comment headers in the source code (or the docbook generated
from them) for more information.

However, given that there are no fewer than four families of RCU APIs
in the Linux kernel, how do you choose which one to use? The following
list can be helpful:

a. Will readers need to block? If so, you need SRCU.

b. What about the -rt patchset? If readers would need to block
in an non-rt kernel, you need SRCU. If readers would block
in a -rt kernel, but not in a non-rt kernel, SRCU is not
necessary.

c. Do you need to treat NMI handlers, hardirq handlers,
and code segments with preemption disabled (whether
via preempt_disable(), local_irq_save(), local_bh_disable(),
or some other mechanism) as if they were explicit RCU readers?
If so, RCU-sched is the only choice that will work for you.

d. Do you need RCU grace periods to complete even in the face
of softirq monopolization of one or more of the CPUs? For
example, is your code subject to network-based denial-of-service
attacks? If so, you need RCU-bh.

e. Is your workload too update-intensive for normal use of
RCU, but inappropriate for other synchronization mechanisms?
If so, consider SLAB_DESTROY_BY_RCU. But please be careful!

f. Do you need read-side critical sections that are respected
even though they are in the middle of the idle loop, during
user-mode execution, or on an offlined CPU? If so, SRCU is the
only choice that will work for you.

g. Otherwise, use RCU.

Of course, this all assumes that you have determined that RCU is in fact
the right tool for your job.

8. ANSWERS TO QUICK QUIZZES

Quick Quiz #1: Why is this argument naive? How could a deadlock
occur when using this algorithm in a real-world Linux
kernel? [Referring to the lock-based "toy" RCU
algorithm.]

Answer: Consider the following sequence of events:

    1.    CPU 0 acquires some unrelated lock, call it  
        "problematic\_lock", disabling irq via  
        spin\_lock\_irqsave().

    2.    CPU 1 enters synchronize\_rcu(), write-acquiring  
        rcu\_gp\_mutex.

    3.    CPU 0 enters rcu\_read\_lock(), but must wait  
        because CPU 1 holds rcu\_gp\_mutex.

    4.    CPU 1 is interrupted, and the irq handler  
        attempts to acquire problematic\_lock.

    The system is now deadlocked.

    One way to avoid this deadlock is to use an approach like  
    that of CONFIG\_PREEMPT\_RT, where all normal spinlocks  
    become blocking locks, and all irq handlers execute in  
    the context of special tasks.  In this case, in step 4  
    above, the irq handler would block, allowing CPU 1 to  
    release rcu\_gp\_mutex, avoiding the deadlock.

    Even in the absence of deadlock, this RCU implementation  
    allows latency to "bleed" from readers to other  
    readers through synchronize\_rcu().  To see this,  
    consider task A in an RCU read-side critical section  
    (thus read-holding rcu\_gp\_mutex), task B blocked  
    attempting to write-acquire rcu\_gp\_mutex, and  
    task C blocked in rcu\_read\_lock() attempting to  
    read\_acquire rcu\_gp\_mutex.  Task A's RCU read-side  
    latency is holding up task C, albeit indirectly via  
    task B.

    Realtime RCU implementations therefore use a counter-based  
    approach where tasks in RCU read-side critical sections  
    cannot be blocked by tasks executing synchronize\_rcu().

Quick Quiz #2: Give an example where Classic RCU's read-side
overhead is -negative-.

Answer: Imagine a single-CPU system with a non-CONFIG_PREEMPT
kernel where a routing table is used by process-context
code, but can be updated by irq-context code (for example,
by an "ICMP REDIRECT" packet). The usual way of handling
this would be to have the process-context code disable
interrupts while searching the routing table. Use of
RCU allows such interrupt-disabling to be dispensed with.
Thus, without RCU, you pay the cost of disabling interrupts,
and with RCU you don't.

    One can argue that the overhead of RCU in this  
    case is negative with respect to the single-CPU  
    interrupt-disabling approach.  Others might argue that  
    the overhead of RCU is merely zero, and that replacing  
    the positive overhead of the interrupt-disabling scheme  
    with the zero-overhead RCU scheme does not constitute  
    negative overhead.

    In real life, of course, things are more complex.  But  
    even the theoretical possibility of negative overhead for  
    a synchronization primitive is a bit unexpected.  ;-)

Quick Quiz #3: If it is illegal to block in an RCU read-side
critical section, what the heck do you do in
PREEMPT_RT, where normal spinlocks can block???

Answer: Just as PREEMPT_RT permits preemption of spinlock
critical sections, it permits preemption of RCU
read-side critical sections. It also permits
spinlocks blocking while in RCU read-side critical
sections.

    Why the apparent inconsistency?  Because it is it  
    possible to use priority boosting to keep the RCU  
    grace periods short if need be (for example, if running  
    short of memory).  In contrast, if blocking waiting  
    for (say) network reception, there is no way to know  
    what should be boosted.  Especially given that the  
    process we need to boost might well be a human being  
    who just went out for a pizza or something.  And although  
    a computer-operated cattle prod might arouse serious  
    interest, it might also provoke serious objections.  
    Besides, how does the computer know what pizza parlor  
    the human being went to???

ACKNOWLEDGEMENTS

My thanks to the people who helped make this human-readable, including
Jon Walpole, Josh Triplett, Serge Hallyn, Suzanne Wood, and Alan Stern.

For more information, see http://www.rdrop.com/users/paulmck/RCU.

*/

Using RCU to Protect Read-Mostly Linked Lists

One of the best applications of RCU is to protect read-mostly linked lists
("struct list_head" in list.h). One big advantage of this approach
is that all of the required memory barriers are included for you in
the list macros. This document describes several applications of RCU,
with the best fits first.

Example 1: Read-Side Action Taken Outside of Lock, No In-Place Updates

The best applications are cases where, if reader-writer locking were
used, the read-side lock would be dropped before taking any action
based on the results of the search. The most celebrated example is
the routing table. Because the routing table is tracking the state of
equipment outside of the computer, it will at times contain stale data.
Therefore, once the route has been computed, there is no need to hold
the routing table static during transmission of the packet. After all,
you can hold the routing table static all you want, but that won't keep
the external Internet from changing, and it is the state of the external
Internet that really matters. In addition, routing entries are typically
added or deleted, rather than being modified in place.

A straightforward example of this use of RCU may be found in the
system-call auditing support. For example, a reader-writer locked
implementation of audit_filter_task() might be as follows:

static enum audit\_state audit\_filter\_task(struct task\_struct \*tsk)  
{  
    struct audit\_entry \*e;  
    enum audit\_state   state;

    read\_lock(&auditsc\_lock);  
    /\* Note: audit\_netlink\_sem held by caller. \*/  
    list\_for\_each\_entry(e, &audit\_tsklist, list) {  
        if (audit\_filter\_rules(tsk, &e->rule, NULL, &state)) {  
            read\_unlock(&auditsc\_lock);  
            return state;  
        }  
    }  
    read\_unlock(&auditsc\_lock);  
    return AUDIT\_BUILD\_CONTEXT;  
}

Here the list is searched under the lock, but the lock is dropped before
the corresponding value is returned. By the time that this value is acted
on, the list may well have been modified. This makes sense, since if
you are turning auditing off, it is OK to audit a few extra system calls.

This means that RCU can be easily applied to the read side, as follows:

static enum audit\_state audit\_filter\_task(struct task\_struct \*tsk)  
{  
    struct audit\_entry \*e;  
    enum audit\_state   state;

    rcu\_read\_lock();  
    /\* Note: audit\_netlink\_sem held by caller. \*/  
    list\_for\_each\_entry\_rcu(e, &audit\_tsklist, list) {  
        if (audit\_filter\_rules(tsk, &e->rule, NULL, &state)) {  
            rcu\_read\_unlock();  
            return state;  
        }  
    }  
    rcu\_read\_unlock();  
    return AUDIT\_BUILD\_CONTEXT;  
}

The read_lock() and read_unlock() calls have become rcu_read_lock()
and rcu_read_unlock(), respectively, and the list_for_each_entry() has
become list_for_each_entry_rcu(). The _rcu() list-traversal primitives
insert the read-side memory barriers that are required on DEC Alpha CPUs.

The changes to the update side are also straightforward. A reader-writer
lock might be used as follows for deletion and insertion:

static inline int audit\_del\_rule(struct audit\_rule \*rule,  
                 struct list\_head \*list)  
{  
    struct audit\_entry  \*e;

    write\_lock(&auditsc\_lock);  
    list\_for\_each\_entry(e, list, list) {  
        if (!audit\_compare\_rule(rule, &e->rule)) {  
            list\_del(&e->list);  
            write\_unlock(&auditsc\_lock);  
            return 0;  
        }  
    }  
    write\_unlock(&auditsc\_lock);  
    return -EFAULT;        /\* No matching rule \*/  
}

static inline int audit\_add\_rule(struct audit\_entry \*entry,  
                 struct list\_head \*list)  
{  
    write\_lock(&auditsc\_lock);  
    if (entry->rule.flags & AUDIT\_PREPEND) {  
        entry->rule.flags &= ~AUDIT\_PREPEND;  
        list\_add(&entry->list, list);  
    } else {  
        list\_add\_tail(&entry->list, list);  
    }  
    write\_unlock(&auditsc\_lock);  
    return 0;  
}

Following are the RCU equivalents for these two functions:

static inline int audit\_del\_rule(struct audit\_rule \*rule,  
                 struct list\_head \*list)  
{  
    struct audit\_entry  \*e;

    /\* Do not use the \_rcu iterator here, since this is the only  
     \* deletion routine. \*/  
    list\_for\_each\_entry(e, list, list) {  
        if (!audit\_compare\_rule(rule, &e->rule)) {  
            list\_del\_rcu(&e->list);  
            call\_rcu(&e->rcu, audit\_free\_rule);  
            return 0;  
        }  
    }  
    return -EFAULT;        /\* No matching rule \*/  
}

static inline int audit\_add\_rule(struct audit\_entry \*entry,  
                 struct list\_head \*list)  
{  
    if (entry->rule.flags & AUDIT\_PREPEND) {  
        entry->rule.flags &= ~AUDIT\_PREPEND;  
        list\_add\_rcu(&entry->list, list);  
    } else {  
        list\_add\_tail\_rcu(&entry->list, list);  
    }  
    return 0;  
}

Normally, the write_lock() and write_unlock() would be replaced by
a spin_lock() and a spin_unlock(), but in this case, all callers hold
audit_netlink_sem, so no additional locking is required. The auditsc_lock
can therefore be eliminated, since use of RCU eliminates the need for
writers to exclude readers. Normally, the write_lock() calls would
be converted into spin_lock() calls.

The list_del(), list_add(), and list_add_tail() primitives have been
replaced by list_del_rcu(), list_add_rcu(), and list_add_tail_rcu().
The _rcu() list-manipulation primitives add memory barriers that are
needed on weakly ordered CPUs (most of them!). The list_del_rcu()
primitive omits the pointer poisoning debug-assist code that would
otherwise cause concurrent readers to fail spectacularly.

So, when readers can tolerate stale data and when entries are either added
or deleted, without in-place modification, it is very easy to use RCU!

Example 2: Handling In-Place Updates

The system-call auditing code does not update auditing rules in place.
However, if it did, reader-writer-locked code to do so might look as
follows (presumably, the field_count is only permitted to decrease,
otherwise, the added fields would need to be filled in):

static inline int audit\_upd\_rule(struct audit\_rule \*rule,  
                 struct list\_head \*list,  
                 \_\_u32 newaction,  
                 \_\_u32 newfield\_count)  
{  
    struct audit\_entry  \*e;  
    struct audit\_newentry \*ne;

    write\_lock(&auditsc\_lock);  
    /\* Note: audit\_netlink\_sem held by caller. \*/  
    list\_for\_each\_entry(e, list, list) {  
        if (!audit\_compare\_rule(rule, &e->rule)) {  
            e->rule.action = newaction;  
            e->rule.file\_count = newfield\_count;  
            write\_unlock(&auditsc\_lock);  
            return 0;  
        }  
    }  
    write\_unlock(&auditsc\_lock);  
    return -EFAULT;        /\* No matching rule \*/  
}

The RCU version creates a copy, updates the copy, then replaces the old
entry with the newly updated entry. This sequence of actions, allowing
concurrent reads while doing a copy to perform an update, is what gives
RCU ("read-copy update") its name. The RCU code is as follows:

static inline int audit\_upd\_rule(struct audit\_rule \*rule,  
                 struct list\_head \*list,  
                 \_\_u32 newaction,  
                 \_\_u32 newfield\_count)  
{  
    struct audit\_entry  \*e;  
    struct audit\_newentry \*ne;

    list\_for\_each\_entry(e, list, list) {  
        if (!audit\_compare\_rule(rule, &e->rule)) {  
            ne = kmalloc(sizeof(\*entry), GFP\_ATOMIC);  
            if (ne == NULL)  
                return -ENOMEM;  
            audit\_copy\_rule(&ne->rule, &e->rule);  
            ne->rule.action = newaction;  
            ne->rule.file\_count = newfield\_count;  
            list\_replace\_rcu(&e->list, &ne->list);  
            call\_rcu(&e->rcu, audit\_free\_rule);  
            return 0;  
        }  
    }  
    return -EFAULT;        /\* No matching rule \*/  
}

Again, this assumes that the caller holds audit_netlink_sem. Normally,
the reader-writer lock would become a spinlock in this sort of code.

Example 3: Eliminating Stale Data

The auditing examples above tolerate stale data, as do most algorithms
that are tracking external state. Because there is a delay from the
time the external state changes before Linux becomes aware of the change,
additional RCU-induced staleness is normally not a problem.

However, there are many examples where stale data cannot be tolerated.
One example in the Linux kernel is the System V IPC (see the ipc_lock()
function in ipc/util.c). This code checks a "deleted" flag under a
per-entry spinlock, and, if the "deleted" flag is set, pretends that the
entry does not exist. For this to be helpful, the search function must
return holding the per-entry spinlock, as ipc_lock() does in fact do.

Quick Quiz: Why does the search function need to return holding the
per-entry lock for this deleted-flag technique to be helpful?

If the system-call audit module were to ever need to reject stale data,
one way to accomplish this would be to add a "deleted" flag and a "lock"
spinlock to the audit_entry structure, and modify audit_filter_task()
as follows:

static enum audit\_state audit\_filter\_task(struct task\_struct \*tsk)  
{  
    struct audit\_entry \*e;  
    enum audit\_state   state;

    rcu\_read\_lock();  
    list\_for\_each\_entry\_rcu(e, &audit\_tsklist, list) {  
        if (audit\_filter\_rules(tsk, &e->rule, NULL, &state)) {  
            spin\_lock(&e->lock);  
            if (e->deleted) {  
                spin\_unlock(&e->lock);  
                rcu\_read\_unlock();  
                return AUDIT\_BUILD\_CONTEXT;  
            }  
            rcu\_read\_unlock();  
            return state;  
        }  
    }  
    rcu\_read\_unlock();  
    return AUDIT\_BUILD\_CONTEXT;  
}

Note that this example assumes that entries are only added and deleted.
Additional mechanism is required to deal correctly with the
update-in-place performed by audit_upd_rule(). For one thing,
audit_upd_rule() would need additional memory barriers to ensure
that the list_add_rcu() was really executed before the list_del_rcu().

The audit_del_rule() function would need to set the "deleted"
flag under the spinlock as follows:

static inline int audit\_del\_rule(struct audit\_rule \*rule,  
                 struct list\_head \*list)  
{  
    struct audit\_entry  \*e;

    /\* Do not need to use the \_rcu iterator here, since this  
     \* is the only deletion routine. \*/  
    list\_for\_each\_entry(e, list, list) {  
        if (!audit\_compare\_rule(rule, &e->rule)) {  
            spin\_lock(&e->lock);  
            list\_del\_rcu(&e->list);  
            e->deleted = 1;  
            spin\_unlock(&e->lock);  
            call\_rcu(&e->rcu, audit\_free\_rule);  
            return 0;  
        }  
    }  
    return -EFAULT;        /\* No matching rule \*/  
}

Summary

Read-mostly list-based data structures that can tolerate stale data are
the most amenable to use of RCU. The simplest case is where entries are
either added or deleted from the data structure (or atomically modified
in place), but non-atomic in-place modifications can be handled by making
a copy, updating the copy, then replacing the original with the copy.
If stale data cannot be tolerated, then a "deleted" flag may be used
in conjunction with a per-entry spinlock in order to allow the search
function to reject newly deleted data.

Answer to Quick Quiz
Why does the search function need to return holding the per-entry
lock for this deleted-flag technique to be helpful?

If the search function drops the per-entry lock before returning,  
then the caller will be processing stale data in any case.  If it  
is really OK to be processing stale data, then you don't need a  
"deleted" flag.  If processing stale data really is a problem,  
then you need to hold the per-entry lock across all of the code  
that uses the value that was returned.

　　在使用RCU时，对共享资源的访问在大部分时间应该是只读的，写访问应该相对较少，因为写访问多了必然相对于其他锁机制而已更占系统资源，影响效率。其次是读者在持有rcu_read_lock（RCU读锁定函数）的时候，不能发生进程上下文切换，否则，因为写者需要等待读者完成方可进行，则此时写者进程也会一直被阻塞，影响系统的正常运行。再次写者执行完毕后需要调用回调函数，此时发生上下文切换，当前进程进入睡眠，则系统将一直不能调用回调函数，更槽糕的是，此时其它进程若再去执行共享的临界区，必然造成一定的错误。最后一点是受RCU机制保护的资源必须是通过指针访问。因为从RCU机制上看，几乎所有操作都是针对指针数据的;

　　同步函数最为重要，即synchronize_rcu()。读者函数的实质其实很简单：禁止抢占，也就是说在RCU期间不允许发生进程上下文切换，原因上述已提及，即是写者需要等待读者完成方可进行，则此时写者进程也会一直被阻塞，影响系统的正常运行等，故而不允许在RCU期间发生进程上下文切换

　　关于写者函数，主要就是call_rcu和call_rcu_bh两个函数。其中call_rcu能实现的功能是它不会使写者阻塞，因而它可在中断上下文及软中断使用，该函数将函数func挂接到RCU的回调函数链表上，然后立即返回，读者函数中提及的synchronize_rcu()函数在实现时也调用了该函数。而call_rcu_bh函数实现的功能几乎与call_rcu完全相同，唯一的差别是它将软中断的完成当作经历一个quiescent state（静默状态，本节一开始有提及这个概念）， 因此若写者使用了该函数，那么读者需对应的使用rcu_read_lock_bh() 和rcu_read_unlock_bh()。

·　　使用rcu_read_lock_bh() 和rcu_read_unlock_bh()函数的原因是由于call_rcu_bh函数不会使写者阻塞，可在中断上下文及软中断使用。这表明此时系统中的中断和软中断并没有被关闭。那么写者在调用call_rcu_bh函数访问临界区时，RCU机制下的读者也能访问临界区。此时对于读者而言，它若是需要读取临界区的内容，它必须把软中断关闭，以免读者在当前的进程上下文过程中被软中断打断（上述内容提过软中断可以打断当前的进程上下文）。而rcu_read_lock_bh() 和rcu_read_unlock_bh()函数的实质是调用local_bh_disable()和local_bh_enable()函数，显然这是实现了禁止软中断和使能软中断的功能。

　　另外在Linux源码中关于call_rcu_bh函数的注释中还明确说明了如果当前的进程是在中断上下文中，则需要执行rcu_read_lock()和rcu_read_unlock()，结合这两个函数的实现实质表明它实际上禁止或使能内核的抢占调度，原因不言而喻，避免当前进程在执行读写过程中被其它进程抢占。同时内核注释还表明call_rcu_bh这个接口函数的使用条件是在大部分的读临界区操作发生在软中断上下文中，原因还是需从它实现的功能出发，相信很容易理解，主要是要从执行效率方面考虑。

https://www.cnblogs.com/alantu2018/p/8459359.html