microsoft
/
WSL2-Linux-Kernel
зеркало из https://github.com/microsoft/WSL2-Linux-Kernel.git
1
0
Форкнуть 0
Код Задачи Пакеты Проекты Релизы Вики Активность

RCU pull request for v6.1 

This pull request contains the following branches:
 
 doc.2022.08.31b: Documentation updates.  This is the first in a series
 	from an ongoing review of the RCU documentation.  "Why are people
 	thinking -that- about RCU?  Oh.  Because that is an entirely
 	reasonable interpretation of its documentation."
 
 fixes.2022.08.31b: Miscellaneous fixes.
 
 kvfree.2022.08.31b: Improved memory allocation and heuristics.
 
 nocb.2022.09.01a: Improve rcu_nocbs diagnostic output.
 
 poll.2022.08.31b: Add full-sized polled RCU grace period state values.
 	These are the same size as an rcu_head structure, which is double
 	that of the traditional unsigned long state values that may still
 	be obtained from et_state_synchronize_rcu().  The added size
 	avoids missing overlapping grace periods.  This benefit is that
 	call_rcu() can be replaced by polling, which can be attractive
 	in situations where RCU-protected data is aged out of memory.
 
 	Early in the series, the size of this state value is three
 	unsigned longs.  Later in the series, the synchronize_rcu() and
 	synchronize_rcu_expedited() fastpaths are reworked to permit
 	the full state to be represented by only two unsigned longs.
 	This reworking slows these two functions down in SMP kernels
 	running either on single-CPU systems or on systems with all but
 	one CPU offlined, but this should not be a significant problem.
 	And if it somehow becomes a problem in some yet-as-unforeseen
 	situations, three-value state values can be provided for only
 	those situations.
 
 	Finally, a pair of functions named same_state_synchronize_rcu()
 	and same_state_synchronize_rcu_full() allow grace-period state
 	values to be compared for equality.  This permits users to
 	maintain lists of data structures having the same state value,
 	removing the need for per-data-structure grace-period state
 	values, thus decreasing memory footprint.
 
 poll-srcu.2022.08.31b: Polled SRCU grace-period updates, including
 	adding tests to rcutorture and reducing the incidence of Tiny
 	SRCU grace-period-state counter wrap.
 
 tasks.2022.08.31b: Improve Tasks RCU diagnostics and quiescent-state
 	detection.
 -----BEGIN PGP SIGNATURE-----
 
 iQJHBAABCgAxFiEEbK7UrM+RBIrCoViJnr8S83LZ+4wFAmM3bxQTHHBhdWxtY2tA
 a2VybmVsLm9yZwAKCRCevxLzctn7jC0zD/0cPe9Nl3LPKVTqDN8wWG6SUOHcwQrg
 dLBUo1pbBh3mK3HcHzwl1iIF7gd2nmKN7UT3m+C+qv+N3Q9ej9K+MutGThCiRvNT
 A56TDYU9I1xfqoQ25E9TL7nqty818rtYYMl36Rw8epcLKHo/It9MFODb5kEBY5ir
 P5UaIK2D4heHfJL6Di8JDq9vC5a/NlNIIkiIj7lUB+px0FpVW0dUqmnWbIOE74YH
 OBGJ/Mxn6KDO4WeFO0v0DxVaBTLd+khu6W0JspI0szOO6iyTqiDCGE5EqkEdcs5I
 Fk9WCifdo9nrQG0LPIuEBv0YnwNfGbe5nYXupAmGFb3tdCbjkM+W0UBUE032nXog
 3E6m5FEBD1XGQttScFHm70kYssa+xI7khGb9/ZFoYN/QW28oWqwfnx6+eAZGxPNS
 AZx6pc2bebg8sOUhkz/Sv+qMH7CQgIgcMR66SKl5SdT1Onaig45sgdUuC23BshgG
 oEdDxvK7vexFQT6q0oqU8LAO/CVKdyVIswt3pB6CUmn8yNgSo+qDZzlEHt0gPdMY
 4Xa1jnNtOHobDnI4g0JMdVqAujByrRq74ZsVW96hdedKrA0r9y462jnVBm9tqW68
 lu0Lw9WLif2kw0lMY8Q59zqTL+fB8TdNiZrHoqefwvQ/ZrvinfHGSblcrS8zAhX3
 4oVwCUs9pPQRMA==
 =AZPZ
 -----END PGP SIGNATURE-----

Merge tag 'rcu.2022.09.30a' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu

Pull RCU updates from Paul McKenney:

 - Documentation updates.

   This is the first in a series from an ongoing review of the RCU
   documentation. "Why are people thinking -that- about RCU? Oh. Because
   that is an entirely reasonable interpretation of its documentation."

 - Miscellaneous fixes.

 - Improved memory allocation and heuristics.

 - Improve rcu_nocbs diagnostic output.

 - Add full-sized polled RCU grace period state values.

   These are the same size as an rcu_head structure, which is double
   that of the traditional unsigned long state values that may still be
   obtained from et_state_synchronize_rcu(). The added size avoids
   missing overlapping grace periods. This benefit is that call_rcu()
   can be replaced by polling, which can be attractive in situations
   where RCU-protected data is aged out of memory.

   Early in the series, the size of this state value is three unsigned
   longs. Later in the series, the fastpaths in synchronize_rcu() and
   synchronize_rcu_expedited() are reworked to permit the full state to
   be represented by only two unsigned longs. This reworking slows these
   two functions down in SMP kernels running either on single-CPU
   systems or on systems with all but one CPU offlined, but this should
   not be a significant problem. And if it somehow becomes a problem in
   some yet-as-unforeseen situations, three-value state values can be
   provided for only those situations.

   Finally, a pair of functions named same_state_synchronize_rcu() and
   same_state_synchronize_rcu_full() allow grace-period state values to
   be compared for equality. This permits users to maintain lists of
   data structures having the same state value, removing the need for
   per-data-structure grace-period state values, thus decreasing memory
   footprint.

 - Polled SRCU grace-period updates, including adding tests to
   rcutorture and reducing the incidence of Tiny SRCU grace-period-state
   counter wrap.

 - Improve Tasks RCU diagnostics and quiescent-state detection.

* tag 'rcu.2022.09.30a' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu: (55 commits)
  rcutorture: Use the barrier operation specified by cur_ops
  rcu-tasks: Make RCU Tasks Trace check for userspace execution
  rcu-tasks: Ensure RCU Tasks Trace loops have quiescent states
  rcu-tasks: Convert RCU_LOCKDEP_WARN() to WARN_ONCE()
  srcu: Make Tiny SRCU use full-sized grace-period counters
  srcu: Make Tiny SRCU poll_state_synchronize_srcu() more precise
  srcu: Add GP and maximum requested GP to Tiny SRCU rcutorture output
  rcutorture: Make "srcud" option also test polled grace-period API
  rcutorture: Limit read-side polling-API testing
  rcu: Add functions to compare grace-period state values
  rcutorture: Expand rcu_torture_write_types() first "if" statement
  rcutorture: Use 1-suffixed variable in rcu_torture_write_types() check
  rcu: Make synchronize_rcu() fastpath update only boot-CPU counters
  rcutorture: Adjust rcu_poll_need_2gp() for rcu_gp_oldstate field removal
  rcu: Remove ->rgos_polled field from rcu_gp_oldstate structure
  rcu: Make synchronize_rcu_expedited() fast path update .expedited_sequence
  rcu: Remove expedited grace-period fast-path forward-progress helper
  rcu: Make synchronize_rcu() fast path update ->gp_seq counters
  rcu-tasks: Remove grace-period fast-path rcu-tasks helper
  rcu: Set rcu_data structures' initial ->gpwrap value to true
  ...
This commit is contained in:
Linus Torvalds 2022-10-03 10:11:11 -07:00
Родитель b8fb65e1d3 5c0ec49004
Коммит 890f242084
18 изменённых файлов: 813 добавлений и 186 удалений
Показать статистику Скачать Patch файл Скачать Diff файл Показать все файлы Свернуть все файлы

15
Documentation/RCU/checklist.rst
Просмотреть файл

@ -66,8 +66,13 @@ over a rather long period of time, but improvements are always welcome!
As a rough rule of thumb, any dereference of an RCU-protected
pointer must be covered by rcu_read_lock(), rcu_read_lock_bh(),
rcu_read_lock_sched(), or by the appropriate update-side lock.
Disabling of preemption can serve as rcu_read_lock_sched(), but
is less readable and prevents lockdep from detecting locking issues.
Explicit disabling of preemption (preempt_disable(), for example)
can serve as rcu_read_lock_sched(), but is less readable and
prevents lockdep from detecting locking issues.
Please not that you *cannot* rely on code known to be built
only in non-preemptible kernels. Such code can and will break,
especially in kernels built with CONFIG_PREEMPT_COUNT=y.
Letting RCU-protected pointers "leak" out of an RCU read-side
critical section is every bit as bad as letting them leak out
@ -185,6 +190,9 @@ over a rather long period of time, but improvements are always welcome!
5. If call_rcu() or call_srcu() is used, the callback function will
be called from softirq context. In particular, it cannot block.
If you need the callback to block, run that code in a workqueue
handler scheduled from the callback. The queue_rcu_work()
function does this for you in the case of call_rcu().
6. Since synchronize_rcu() can block, it cannot be called
from any sort of irq context. The same rule applies
@ -297,7 +305,8 @@ over a rather long period of time, but improvements are always welcome!
the machine.
d. Periodically invoke synchronize_rcu(), permitting a limited
number of updates per grace period.
number of updates per grace period. Better yet, periodically
invoke rcu_barrier() to wait for all outstanding callbacks.
The same cautions apply to call_srcu() and kfree_rcu().

14
Documentation/RCU/rcu_dereference.rst
Просмотреть файл

@ -128,10 +128,16 @@ Follow these rules to keep your RCU code working properly:
This sort of comparison occurs frequently when scanning
RCU-protected circular linked lists.
Note that if checks for being within an RCU read-side
critical section are not required and the pointer is never
dereferenced, rcu_access_pointer() should be used in place
of rcu_dereference().
Note that if the pointer comparison is done outside
of an RCU read-side critical section, and the pointer
is never dereferenced, rcu_access_pointer() should be
used in place of rcu_dereference(). In most cases,
it is best to avoid accidental dereferences by testing
the rcu_access_pointer() return value directly, without
assigning it to a variable.
Within an RCU read-side critical section, there is little
reason to use rcu_access_pointer().
- The comparison is against a pointer that references memory
that was initialized "a long time ago." The reason

47
Documentation/RCU/whatisRCU.rst
Просмотреть файл

@ -6,13 +6,15 @@ What is RCU? -- "Read, Copy, Update"
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/
| 2010 Big API Table http://lwn.net/Articles/419086/
| 5. The RCU API, 2014 Edition http://lwn.net/Articles/609904/
| 2014 Big API Table http://lwn.net/Articles/609973/
| 1. What is RCU, Fundamentally? https://lwn.net/Articles/262464/
| 2. What is RCU? Part 2: Usage https://lwn.net/Articles/263130/
| 3. RCU part 3: the RCU API https://lwn.net/Articles/264090/
| 4. The RCU API, 2010 Edition https://lwn.net/Articles/418853/
| 2010 Big API Table https://lwn.net/Articles/419086/
| 5. The RCU API, 2014 Edition https://lwn.net/Articles/609904/
| 2014 Big API Table https://lwn.net/Articles/609973/
| 6. The RCU API, 2019 Edition https://lwn.net/Articles/777036/
| 2019 Big API Table https://lwn.net/Articles/777165/
What is RCU?
@ -915,13 +917,18 @@ which an RCU reference is held include:
The understanding that RCU provides a reference that only prevents a
change of type is particularly visible with objects allocated from a
slab cache marked ``SLAB_TYPESAFE_BY_RCU``. RCU operations may yield a
reference to an object from such a cache that has been concurrently
freed and the memory reallocated to a completely different object,
though of the same type. In this case RCU doesn't even protect the
identity of the object from changing, only its type. So the object
found may not be the one expected, but it will be one where it is safe
to take a reference or spinlock and then confirm that the identity
matches the expectations.
reference to an object from such a cache that has been concurrently freed
and the memory reallocated to a completely different object, though of
the same type. In this case RCU doesn't even protect the identity of the
object from changing, only its type. So the object found may not be the
one expected, but it will be one where it is safe to take a reference
(and then potentially acquiring a spinlock), allowing subsequent code
to check whether the identity matches expectations. It is tempting
to simply acquire the spinlock without first taking the reference, but
unfortunately any spinlock in a ``SLAB_TYPESAFE_BY_RCU`` object must be
initialized after each and every call to kmem_cache_alloc(), which renders
reference-free spinlock acquisition completely unsafe. Therefore, when
using ``SLAB_TYPESAFE_BY_RCU``, make proper use of a reference counter.
With traditional reference counting -- such as that implemented by the
kref library in Linux -- there is typically code that runs when the last
@ -1057,14 +1064,20 @@ SRCU: Initialization/cleanup::
init_srcu_struct
cleanup_srcu_struct
All: lockdep-checked RCU-protected pointer access::
All: lockdep-checked RCU utility APIs::
rcu_access_pointer
rcu_dereference_raw
RCU_LOCKDEP_WARN
rcu_sleep_check
RCU_NONIDLE
All: Unchecked RCU-protected pointer access::
rcu_dereference_raw
All: Unchecked RCU-protected pointer access with dereferencing prohibited::
rcu_access_pointer
See the comment headers in the source code (or the docbook generated
from them) for more information.

42
include/linux/rcupdate.h
Просмотреть файл

@ -42,7 +42,31 @@ void call_rcu(struct rcu_head *head, rcu_callback_t func);
void rcu_barrier_tasks(void);
void rcu_barrier_tasks_rude(void);
void synchronize_rcu(void);
struct rcu_gp_oldstate;
unsigned long get_completed_synchronize_rcu(void);
void get_completed_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp);
// Maximum number of unsigned long values corresponding to
// not-yet-completed RCU grace periods.
#define NUM_ACTIVE_RCU_POLL_OLDSTATE 2
/**
* same_state_synchronize_rcu - Are two old-state values identical?
* @oldstate1: First old-state value.
* @oldstate2: Second old-state value.
*
* The two old-state values must have been obtained from either
* get_state_synchronize_rcu(), start_poll_synchronize_rcu(), or
* get_completed_synchronize_rcu(). Returns @true if the two values are
* identical and @false otherwise. This allows structures whose lifetimes
* are tracked by old-state values to push these values to a list header,
* allowing those structures to be slightly smaller.
*/
static inline bool same_state_synchronize_rcu(unsigned long oldstate1, unsigned long oldstate2)
{
return oldstate1 == oldstate2;
}
#ifdef CONFIG_PREEMPT_RCU
@ -496,13 +520,21 @@ do { \
* against NULL. Although rcu_access_pointer() may also be used in cases
* where update-side locks prevent the value of the pointer from changing,
* you should instead use rcu_dereference_protected() for this use case.
* Within an RCU read-side critical section, there is little reason to
* use rcu_access_pointer().
*
* It is usually best to test the rcu_access_pointer() return value
* directly in order to avoid accidental dereferences being introduced
* by later inattentive changes. In other words, assigning the
* rcu_access_pointer() return value to a local variable results in an
* accident waiting to happen.
*
* It is also permissible to use rcu_access_pointer() when read-side
* access to the pointer was removed at least one grace period ago, as
* is the case in the context of the RCU callback that is freeing up
* the data, or after a synchronize_rcu() returns. This can be useful
* when tearing down multi-linked structures after a grace period
* has elapsed.
* access to the pointer was removed at least one grace period ago, as is
* the case in the context of the RCU callback that is freeing up the data,
* or after a synchronize_rcu() returns. This can be useful when tearing
* down multi-linked structures after a grace period has elapsed. However,
* rcu_dereference_protected() is normally preferred for this use case.
*/
#define rcu_access_pointer(p) __rcu_access_pointer((p), __UNIQUE_ID(rcu), __rcu)

50
include/linux/rcutiny.h
Просмотреть файл

@ -14,25 +14,75 @@
#include <asm/param.h> /* for HZ */
struct rcu_gp_oldstate {
unsigned long rgos_norm;
};
// Maximum number of rcu_gp_oldstate values corresponding to
// not-yet-completed RCU grace periods.
#define NUM_ACTIVE_RCU_POLL_FULL_OLDSTATE 2
/*
* Are the two oldstate values the same? See the Tree RCU version for
* docbook header.
*/
static inline bool same_state_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp1,
struct rcu_gp_oldstate *rgosp2)
{
return rgosp1->rgos_norm == rgosp2->rgos_norm;
}
unsigned long get_state_synchronize_rcu(void);
static inline void get_state_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp)
{
rgosp->rgos_norm = get_state_synchronize_rcu();
}
unsigned long start_poll_synchronize_rcu(void);
static inline void start_poll_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp)
{
rgosp->rgos_norm = start_poll_synchronize_rcu();
}
bool poll_state_synchronize_rcu(unsigned long oldstate);
static inline bool poll_state_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp)
{
return poll_state_synchronize_rcu(rgosp->rgos_norm);
}
static inline void cond_synchronize_rcu(unsigned long oldstate)
{
might_sleep();
}
static inline void cond_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp)
{
cond_synchronize_rcu(rgosp->rgos_norm);
}
static inline unsigned long start_poll_synchronize_rcu_expedited(void)
{
return start_poll_synchronize_rcu();
}
static inline void start_poll_synchronize_rcu_expedited_full(struct rcu_gp_oldstate *rgosp)
{
rgosp->rgos_norm = start_poll_synchronize_rcu_expedited();
}
static inline void cond_synchronize_rcu_expedited(unsigned long oldstate)
{
cond_synchronize_rcu(oldstate);
}
static inline void cond_synchronize_rcu_expedited_full(struct rcu_gp_oldstate *rgosp)
{
cond_synchronize_rcu_expedited(rgosp->rgos_norm);
}
extern void rcu_barrier(void);
static inline void synchronize_rcu_expedited(void)

40
include/linux/rcutree.h
Просмотреть файл

@ -40,12 +40,52 @@ bool rcu_eqs_special_set(int cpu);
void rcu_momentary_dyntick_idle(void);
void kfree_rcu_scheduler_running(void);
bool rcu_gp_might_be_stalled(void);
struct rcu_gp_oldstate {
unsigned long rgos_norm;
unsigned long rgos_exp;
};
// Maximum number of rcu_gp_oldstate values corresponding to
// not-yet-completed RCU grace periods.
#define NUM_ACTIVE_RCU_POLL_FULL_OLDSTATE 4
/**
* same_state_synchronize_rcu_full - Are two old-state values identical?
* @rgosp1: First old-state value.
* @rgosp2: Second old-state value.
*
* The two old-state values must have been obtained from either
* get_state_synchronize_rcu_full(), start_poll_synchronize_rcu_full(),
* or get_completed_synchronize_rcu_full(). Returns @true if the two
* values are identical and @false otherwise. This allows structures
* whose lifetimes are tracked by old-state values to push these values
* to a list header, allowing those structures to be slightly smaller.
*
* Note that equality is judged on a bitwise basis, so that an
* @rcu_gp_oldstate structure with an already-completed state in one field
* will compare not-equal to a structure with an already-completed state
* in the other field. After all, the @rcu_gp_oldstate structure is opaque
* so how did such a situation come to pass in the first place?
*/
static inline bool same_state_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp1,
struct rcu_gp_oldstate *rgosp2)
{
return rgosp1->rgos_norm == rgosp2->rgos_norm && rgosp1->rgos_exp == rgosp2->rgos_exp;
}
unsigned long start_poll_synchronize_rcu_expedited(void);
void start_poll_synchronize_rcu_expedited_full(struct rcu_gp_oldstate *rgosp);
void cond_synchronize_rcu_expedited(unsigned long oldstate);
void cond_synchronize_rcu_expedited_full(struct rcu_gp_oldstate *rgosp);
unsigned long get_state_synchronize_rcu(void);
void get_state_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp);
unsigned long start_poll_synchronize_rcu(void);
void start_poll_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp);
bool poll_state_synchronize_rcu(unsigned long oldstate);
bool poll_state_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp);
void cond_synchronize_rcu(unsigned long oldstate);
void cond_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp);
bool rcu_is_idle_cpu(int cpu);

10
include/linux/srcutiny.h
Просмотреть файл

@ -15,10 +15,10 @@
struct srcu_struct {
short srcu_lock_nesting[2]; /* srcu_read_lock() nesting depth. */
unsigned short srcu_idx; /* Current reader array element in bit 0x2. */
unsigned short srcu_idx_max; /* Furthest future srcu_idx request. */
u8 srcu_gp_running; /* GP workqueue running? */
u8 srcu_gp_waiting; /* GP waiting for readers? */
unsigned long srcu_idx; /* Current reader array element in bit 0x2. */
unsigned long srcu_idx_max; /* Furthest future srcu_idx request. */
struct swait_queue_head srcu_wq;
/* Last srcu_read_unlock() wakes GP. */
struct rcu_head *srcu_cb_head; /* Pending callbacks: Head. */
@ -82,10 +82,12 @@ static inline void srcu_torture_stats_print(struct srcu_struct *ssp,
int idx;
idx = ((data_race(READ_ONCE(ssp->srcu_idx)) + 1) & 0x2) >> 1;
pr_alert("%s%s Tiny SRCU per-CPU(idx=%d): (%hd,%hd)\n",
pr_alert("%s%s Tiny SRCU per-CPU(idx=%d): (%hd,%hd) gp: %lu->%lu\n",
tt, tf, idx,
data_race(READ_ONCE(ssp->srcu_lock_nesting[!idx])),
data_race(READ_ONCE(ssp->srcu_lock_nesting[idx])));
data_race(READ_ONCE(ssp->srcu_lock_nesting[idx])),
data_race(READ_ONCE(ssp->srcu_idx)),
data_race(READ_ONCE(ssp->srcu_idx_max)));
}
#endif

272
kernel/rcu/rcutorture.c
Просмотреть файл

@ -84,10 +84,15 @@ torture_param(int, fwd_progress_holdoff, 60, "Time between forward-progress test
torture_param(bool, fwd_progress_need_resched, 1, "Hide cond_resched() behind need_resched()");
torture_param(bool, gp_cond, false, "Use conditional/async GP wait primitives");
torture_param(bool, gp_cond_exp, false, "Use conditional/async expedited GP wait primitives");
torture_param(bool, gp_cond_full, false, "Use conditional/async full-state GP wait primitives");
torture_param(bool, gp_cond_exp_full, false,
"Use conditional/async full-stateexpedited GP wait primitives");
torture_param(bool, gp_exp, false, "Use expedited GP wait primitives");
torture_param(bool, gp_normal, false, "Use normal (non-expedited) GP wait primitives");
torture_param(bool, gp_poll, false, "Use polling GP wait primitives");
torture_param(bool, gp_poll_exp, false, "Use polling expedited GP wait primitives");
torture_param(bool, gp_poll_full, false, "Use polling full-state GP wait primitives");
torture_param(bool, gp_poll_exp_full, false, "Use polling full-state expedited GP wait primitives");
torture_param(bool, gp_sync, false, "Use synchronous GP wait primitives");
torture_param(int, irqreader, 1, "Allow RCU readers from irq handlers");
torture_param(int, leakpointer, 0, "Leak pointer dereferences from readers");
@ -194,16 +199,24 @@ static int rcu_torture_writer_state;
#define RTWS_DEF_FREE 3
#define RTWS_EXP_SYNC 4
#define RTWS_COND_GET 5
#define RTWS_COND_GET_EXP 6
#define RTWS_COND_SYNC 7
#define RTWS_COND_SYNC_EXP 8
#define RTWS_POLL_GET 9
#define RTWS_POLL_GET_EXP 10
#define RTWS_POLL_WAIT 11
#define RTWS_POLL_WAIT_EXP 12
#define RTWS_SYNC 13
#define RTWS_STUTTER 14
#define RTWS_STOPPING 15
#define RTWS_COND_GET_FULL 6
#define RTWS_COND_GET_EXP 7
#define RTWS_COND_GET_EXP_FULL 8
#define RTWS_COND_SYNC 9
#define RTWS_COND_SYNC_FULL 10
#define RTWS_COND_SYNC_EXP 11
#define RTWS_COND_SYNC_EXP_FULL 12
#define RTWS_POLL_GET 13
#define RTWS_POLL_GET_FULL 14
#define RTWS_POLL_GET_EXP 15
#define RTWS_POLL_GET_EXP_FULL 16
#define RTWS_POLL_WAIT 17
#define RTWS_POLL_WAIT_FULL 18
#define RTWS_POLL_WAIT_EXP 19
#define RTWS_POLL_WAIT_EXP_FULL 20
#define RTWS_SYNC 21
#define RTWS_STUTTER 22
#define RTWS_STOPPING 23
static const char * const rcu_torture_writer_state_names[] = {
"RTWS_FIXED_DELAY",
"RTWS_DELAY",
@ -211,13 +224,21 @@ static const char * const rcu_torture_writer_state_names[] = {
"RTWS_DEF_FREE",
"RTWS_EXP_SYNC",
"RTWS_COND_GET",
"RTWS_COND_GET_FULL",
"RTWS_COND_GET_EXP",
"RTWS_COND_GET_EXP_FULL",
"RTWS_COND_SYNC",
"RTWS_COND_SYNC_FULL",
"RTWS_COND_SYNC_EXP",
"RTWS_COND_SYNC_EXP_FULL",
"RTWS_POLL_GET",
"RTWS_POLL_GET_FULL",
"RTWS_POLL_GET_EXP",
"RTWS_POLL_GET_EXP_FULL",
"RTWS_POLL_WAIT",
"RTWS_POLL_WAIT_FULL",
"RTWS_POLL_WAIT_EXP",
"RTWS_POLL_WAIT_EXP_FULL",
"RTWS_SYNC",
"RTWS_STUTTER",
"RTWS_STOPPING",
@ -332,13 +353,21 @@ struct rcu_torture_ops {
void (*exp_sync)(void);
unsigned long (*get_gp_state_exp)(void);
unsigned long (*start_gp_poll_exp)(void);
void (*start_gp_poll_exp_full)(struct rcu_gp_oldstate *rgosp);
bool (*poll_gp_state_exp)(unsigned long oldstate);
void (*cond_sync_exp)(unsigned long oldstate);
void (*cond_sync_exp_full)(struct rcu_gp_oldstate *rgosp);
unsigned long (*get_gp_state)(void);
void (*get_gp_state_full)(struct rcu_gp_oldstate *rgosp);
unsigned long (*get_gp_completed)(void);
void (*get_gp_completed_full)(struct rcu_gp_oldstate *rgosp);
unsigned long (*start_gp_poll)(void);
void (*start_gp_poll_full)(struct rcu_gp_oldstate *rgosp);
bool (*poll_gp_state)(unsigned long oldstate);
bool (*poll_gp_state_full)(struct rcu_gp_oldstate *rgosp);
bool (*poll_need_2gp)(bool poll, bool poll_full);
void (*cond_sync)(unsigned long oldstate);
void (*cond_sync_full)(struct rcu_gp_oldstate *rgosp);
call_rcu_func_t call;
void (*cb_barrier)(void);
void (*fqs)(void);
@ -489,6 +518,11 @@ static void rcu_sync_torture_init(void)
INIT_LIST_HEAD(&rcu_torture_removed);
}
static bool rcu_poll_need_2gp(bool poll, bool poll_full)
{
return poll;
}
static struct rcu_torture_ops rcu_ops = {
.ttype = RCU_FLAVOR,
.init = rcu_sync_torture_init,
@ -502,12 +536,19 @@ static struct rcu_torture_ops rcu_ops = {
.sync = synchronize_rcu,
.exp_sync = synchronize_rcu_expedited,
.get_gp_state = get_state_synchronize_rcu,
.get_gp_state_full = get_state_synchronize_rcu_full,
.get_gp_completed = get_completed_synchronize_rcu,
.get_gp_completed_full = get_completed_synchronize_rcu_full,
.start_gp_poll = start_poll_synchronize_rcu,
.start_gp_poll_full = start_poll_synchronize_rcu_full,
.poll_gp_state = poll_state_synchronize_rcu,
.poll_gp_state_full = poll_state_synchronize_rcu_full,
.poll_need_2gp = rcu_poll_need_2gp,
.cond_sync = cond_synchronize_rcu,
.cond_sync_full = cond_synchronize_rcu_full,
.get_gp_state_exp = get_state_synchronize_rcu,
.start_gp_poll_exp = start_poll_synchronize_rcu_expedited,
.start_gp_poll_exp_full = start_poll_synchronize_rcu_expedited_full,
.poll_gp_state_exp = poll_state_synchronize_rcu,
.cond_sync_exp = cond_synchronize_rcu_expedited,
.call = call_rcu,
@ -709,6 +750,9 @@ static struct rcu_torture_ops srcud_ops = {
.deferred_free = srcu_torture_deferred_free,
.sync = srcu_torture_synchronize,
.exp_sync = srcu_torture_synchronize_expedited,
.get_gp_state = srcu_torture_get_gp_state,
.start_gp_poll = srcu_torture_start_gp_poll,
.poll_gp_state = srcu_torture_poll_gp_state,
.call = srcu_torture_call,
.cb_barrier = srcu_torture_barrier,
.stats = srcu_torture_stats,
@ -1148,15 +1192,35 @@ static int nsynctypes;
*/
static void rcu_torture_write_types(void)
{
bool gp_cond1 = gp_cond, gp_cond_exp1 = gp_cond_exp, gp_exp1 = gp_exp;
bool gp_poll_exp1 = gp_poll_exp, gp_normal1 = gp_normal, gp_poll1 = gp_poll;
bool gp_sync1 = gp_sync;
bool gp_cond1 = gp_cond, gp_cond_exp1 = gp_cond_exp, gp_cond_full1 = gp_cond_full;
bool gp_cond_exp_full1 = gp_cond_exp_full, gp_exp1 = gp_exp, gp_poll_exp1 = gp_poll_exp;
bool gp_poll_exp_full1 = gp_poll_exp_full, gp_normal1 = gp_normal, gp_poll1 = gp_poll;
bool gp_poll_full1 = gp_poll_full, gp_sync1 = gp_sync;
/* Initialize synctype[] array. If none set, take default. */
if (!gp_cond1 && !gp_cond_exp1 && !gp_exp1 && !gp_poll_exp &&
!gp_normal1 && !gp_poll1 && !gp_sync1)
gp_cond1 = gp_cond_exp1 = gp_exp1 = gp_poll_exp1 =
gp_normal1 = gp_poll1 = gp_sync1 = true;
if (!gp_cond1 &&
!gp_cond_exp1 &&
!gp_cond_full1 &&
!gp_cond_exp_full1 &&
!gp_exp1 &&
!gp_poll_exp1 &&
!gp_poll_exp_full1 &&
!gp_normal1 &&
!gp_poll1 &&
!gp_poll_full1 &&
!gp_sync1) {
gp_cond1 = true;
gp_cond_exp1 = true;
gp_cond_full1 = true;
gp_cond_exp_full1 = true;
gp_exp1 = true;
gp_poll_exp1 = true;
gp_poll_exp_full1 = true;
gp_normal1 = true;
gp_poll1 = true;
gp_poll_full1 = true;
gp_sync1 = true;
}
if (gp_cond1 && cur_ops->get_gp_state && cur_ops->cond_sync) {
synctype[nsynctypes++] = RTWS_COND_GET;
pr_info("%s: Testing conditional GPs.\n", __func__);
@ -1169,6 +1233,19 @@ static void rcu_torture_write_types(void)
} else if (gp_cond_exp && (!cur_ops->get_gp_state_exp || !cur_ops->cond_sync_exp)) {
pr_alert("%s: gp_cond_exp without primitives.\n", __func__);
}
if (gp_cond_full1 && cur_ops->get_gp_state && cur_ops->cond_sync_full) {
synctype[nsynctypes++] = RTWS_COND_GET_FULL;
pr_info("%s: Testing conditional full-state GPs.\n", __func__);
} else if (gp_cond_full && (!cur_ops->get_gp_state || !cur_ops->cond_sync_full)) {
pr_alert("%s: gp_cond_full without primitives.\n", __func__);
}
if (gp_cond_exp_full1 && cur_ops->get_gp_state_exp && cur_ops->cond_sync_exp_full) {
synctype[nsynctypes++] = RTWS_COND_GET_EXP_FULL;
pr_info("%s: Testing conditional full-state expedited GPs.\n", __func__);
} else if (gp_cond_exp_full &&
(!cur_ops->get_gp_state_exp || !cur_ops->cond_sync_exp_full)) {
pr_alert("%s: gp_cond_exp_full without primitives.\n", __func__);
}
if (gp_exp1 && cur_ops->exp_sync) {
synctype[nsynctypes++] = RTWS_EXP_SYNC;
pr_info("%s: Testing expedited GPs.\n", __func__);
@ -1187,12 +1264,25 @@ static void rcu_torture_write_types(void)
} else if (gp_poll && (!cur_ops->start_gp_poll || !cur_ops->poll_gp_state)) {
pr_alert("%s: gp_poll without primitives.\n", __func__);
}
if (gp_poll_full1 && cur_ops->start_gp_poll_full && cur_ops->poll_gp_state_full) {
synctype[nsynctypes++] = RTWS_POLL_GET_FULL;
pr_info("%s: Testing polling full-state GPs.\n", __func__);
} else if (gp_poll_full && (!cur_ops->start_gp_poll_full || !cur_ops->poll_gp_state_full)) {
pr_alert("%s: gp_poll_full without primitives.\n", __func__);
}
if (gp_poll_exp1 && cur_ops->start_gp_poll_exp && cur_ops->poll_gp_state_exp) {
synctype[nsynctypes++] = RTWS_POLL_GET_EXP;
pr_info("%s: Testing polling expedited GPs.\n", __func__);
} else if (gp_poll_exp && (!cur_ops->start_gp_poll_exp || !cur_ops->poll_gp_state_exp)) {
pr_alert("%s: gp_poll_exp without primitives.\n", __func__);
}
if (gp_poll_exp_full1 && cur_ops->start_gp_poll_exp_full && cur_ops->poll_gp_state_full) {
synctype[nsynctypes++] = RTWS_POLL_GET_EXP_FULL;
pr_info("%s: Testing polling full-state expedited GPs.\n", __func__);
} else if (gp_poll_exp_full &&
(!cur_ops->start_gp_poll_exp_full || !cur_ops->poll_gp_state_full)) {
pr_alert("%s: gp_poll_exp_full without primitives.\n", __func__);
}
if (gp_sync1 && cur_ops->sync) {
synctype[nsynctypes++] = RTWS_SYNC;
pr_info("%s: Testing normal GPs.\n", __func__);
@ -1201,6 +1291,40 @@ static void rcu_torture_write_types(void)
}
}
/*
* Do the specified rcu_torture_writer() synchronous grace period,
* while also testing out the polled APIs. Note well that the single-CPU
* grace-period optimizations must be accounted for.
*/
static void do_rtws_sync(struct torture_random_state *trsp, void (*sync)(void))
{
unsigned long cookie;
struct rcu_gp_oldstate cookie_full;
bool dopoll;
bool dopoll_full;
unsigned long r = torture_random(trsp);
dopoll = cur_ops->get_gp_state && cur_ops->poll_gp_state && !(r & 0x300);
dopoll_full = cur_ops->get_gp_state_full && cur_ops->poll_gp_state_full && !(r & 0xc00);
if (dopoll || dopoll_full)
cpus_read_lock();
if (dopoll)
cookie = cur_ops->get_gp_state();
if (dopoll_full)
cur_ops->get_gp_state_full(&cookie_full);
if (cur_ops->poll_need_2gp && cur_ops->poll_need_2gp(dopoll, dopoll_full))
sync();
sync();
WARN_ONCE(dopoll && !cur_ops->poll_gp_state(cookie),
"%s: Cookie check 3 failed %pS() online %*pbl.",
__func__, sync, cpumask_pr_args(cpu_online_mask));
WARN_ONCE(dopoll_full && !cur_ops->poll_gp_state_full(&cookie_full),
"%s: Cookie check 4 failed %pS() online %*pbl",
__func__, sync, cpumask_pr_args(cpu_online_mask));
if (dopoll || dopoll_full)
cpus_read_unlock();
}
/*
* RCU torture writer kthread. Repeatedly substitutes a new structure
* for that pointed to by rcu_torture_current, freeing the old structure
@ -1212,8 +1336,10 @@ rcu_torture_writer(void *arg)
bool boot_ended;
bool can_expedite = !rcu_gp_is_expedited() && !rcu_gp_is_normal();
unsigned long cookie;
struct rcu_gp_oldstate cookie_full;
int expediting = 0;
unsigned long gp_snap;
struct rcu_gp_oldstate gp_snap_full;
int i;
int idx;
int oldnice = task_nice(current);
@ -1261,11 +1387,12 @@ rcu_torture_writer(void *arg)
atomic_inc(&rcu_torture_wcount[i]);
WRITE_ONCE(old_rp->rtort_pipe_count,
old_rp->rtort_pipe_count + 1);
// Make sure readers block polled grace periods.
if (cur_ops->get_gp_state && cur_ops->poll_gp_state) {
idx = cur_ops->readlock();
cookie = cur_ops->get_gp_state();
WARN_ONCE(rcu_torture_writer_state != RTWS_DEF_FREE &&
cur_ops->poll_gp_state(cookie),
WARN_ONCE(cur_ops->poll_gp_state(cookie),
"%s: Cookie check 1 failed %s(%d) %lu->%lu\n",
__func__,
rcu_torture_writer_state_getname(),
@ -1277,6 +1404,21 @@ rcu_torture_writer(void *arg)
}
cur_ops->readunlock(idx);
}
if (cur_ops->get_gp_state_full && cur_ops->poll_gp_state_full) {
idx = cur_ops->readlock();
cur_ops->get_gp_state_full(&cookie_full);
WARN_ONCE(cur_ops->poll_gp_state_full(&cookie_full),
"%s: Cookie check 5 failed %s(%d) online %*pbl\n",
__func__,
rcu_torture_writer_state_getname(),
rcu_torture_writer_state,
cpumask_pr_args(cpu_online_mask));
if (cur_ops->get_gp_completed_full) {
cur_ops->get_gp_completed_full(&cookie_full);
WARN_ON_ONCE(!cur_ops->poll_gp_state_full(&cookie_full));
}
cur_ops->readunlock(idx);
}
switch (synctype[torture_random(&rand) % nsynctypes]) {
case RTWS_DEF_FREE:
rcu_torture_writer_state = RTWS_DEF_FREE;
@ -1284,12 +1426,7 @@ rcu_torture_writer(void *arg)
break;
case RTWS_EXP_SYNC:
rcu_torture_writer_state = RTWS_EXP_SYNC;
if (cur_ops->get_gp_state && cur_ops->poll_gp_state)
cookie = cur_ops->get_gp_state();
cur_ops->exp_sync();
cur_ops->exp_sync();
if (cur_ops->get_gp_state && cur_ops->poll_gp_state)
WARN_ON_ONCE(!cur_ops->poll_gp_state(cookie));
do_rtws_sync(&rand, cur_ops->exp_sync);
rcu_torture_pipe_update(old_rp);
break;
case RTWS_COND_GET:
@ -1308,6 +1445,22 @@ rcu_torture_writer(void *arg)
cur_ops->cond_sync_exp(gp_snap);
rcu_torture_pipe_update(old_rp);
break;
case RTWS_COND_GET_FULL:
rcu_torture_writer_state = RTWS_COND_GET_FULL;
cur_ops->get_gp_state_full(&gp_snap_full);
torture_hrtimeout_jiffies(torture_random(&rand) % 16, &rand);
rcu_torture_writer_state = RTWS_COND_SYNC_FULL;
cur_ops->cond_sync_full(&gp_snap_full);
rcu_torture_pipe_update(old_rp);
break;
case RTWS_COND_GET_EXP_FULL:
rcu_torture_writer_state = RTWS_COND_GET_EXP_FULL;
cur_ops->get_gp_state_full(&gp_snap_full);
torture_hrtimeout_jiffies(torture_random(&rand) % 16, &rand);
rcu_torture_writer_state = RTWS_COND_SYNC_EXP_FULL;
cur_ops->cond_sync_exp_full(&gp_snap_full);
rcu_torture_pipe_update(old_rp);
break;
case RTWS_POLL_GET:
rcu_torture_writer_state = RTWS_POLL_GET;
gp_snap = cur_ops->start_gp_poll();
@ -1317,6 +1470,15 @@ rcu_torture_writer(void *arg)
&rand);
rcu_torture_pipe_update(old_rp);
break;
case RTWS_POLL_GET_FULL:
rcu_torture_writer_state = RTWS_POLL_GET_FULL;
cur_ops->start_gp_poll_full(&gp_snap_full);
rcu_torture_writer_state = RTWS_POLL_WAIT_FULL;
while (!cur_ops->poll_gp_state_full(&gp_snap_full))
torture_hrtimeout_jiffies(torture_random(&rand) % 16,
&rand);
rcu_torture_pipe_update(old_rp);
break;
case RTWS_POLL_GET_EXP:
rcu_torture_writer_state = RTWS_POLL_GET_EXP;
gp_snap = cur_ops->start_gp_poll_exp();
@ -1326,14 +1488,18 @@ rcu_torture_writer(void *arg)
&rand);
rcu_torture_pipe_update(old_rp);
break;
case RTWS_POLL_GET_EXP_FULL:
rcu_torture_writer_state = RTWS_POLL_GET_EXP_FULL;
cur_ops->start_gp_poll_exp_full(&gp_snap_full);
rcu_torture_writer_state = RTWS_POLL_WAIT_EXP_FULL;
while (!cur_ops->poll_gp_state_full(&gp_snap_full))
torture_hrtimeout_jiffies(torture_random(&rand) % 16,
&rand);
rcu_torture_pipe_update(old_rp);
break;
case RTWS_SYNC:
rcu_torture_writer_state = RTWS_SYNC;
if (cur_ops->get_gp_state && cur_ops->poll_gp_state)
cookie = cur_ops->get_gp_state();
cur_ops->sync();
cur_ops->sync();
if (cur_ops->get_gp_state && cur_ops->poll_gp_state)
WARN_ON_ONCE(!cur_ops->poll_gp_state(cookie));
do_rtws_sync(&rand, cur_ops->sync);
rcu_torture_pipe_update(old_rp);
break;
default:
@ -1400,6 +1566,7 @@ static int
rcu_torture_fakewriter(void *arg)
{
unsigned long gp_snap;
struct rcu_gp_oldstate gp_snap_full;
DEFINE_TORTURE_RANDOM(rand);
VERBOSE_TOROUT_STRING("rcu_torture_fakewriter task started");
@ -1438,6 +1605,16 @@ rcu_torture_fakewriter(void *arg)
torture_hrtimeout_jiffies(torture_random(&rand) % 16, &rand);
cur_ops->cond_sync_exp(gp_snap);
break;
case RTWS_COND_GET_FULL:
cur_ops->get_gp_state_full(&gp_snap_full);
torture_hrtimeout_jiffies(torture_random(&rand) % 16, &rand);
cur_ops->cond_sync_full(&gp_snap_full);
break;
case RTWS_COND_GET_EXP_FULL:
cur_ops->get_gp_state_full(&gp_snap_full);
torture_hrtimeout_jiffies(torture_random(&rand) % 16, &rand);
cur_ops->cond_sync_exp_full(&gp_snap_full);
break;
case RTWS_POLL_GET:
gp_snap = cur_ops->start_gp_poll();
while (!cur_ops->poll_gp_state(gp_snap)) {
@ -1445,6 +1622,13 @@ rcu_torture_fakewriter(void *arg)
&rand);
}
break;
case RTWS_POLL_GET_FULL:
cur_ops->start_gp_poll_full(&gp_snap_full);
while (!cur_ops->poll_gp_state_full(&gp_snap_full)) {
torture_hrtimeout_jiffies(torture_random(&rand) % 16,
&rand);
}
break;
case RTWS_POLL_GET_EXP:
gp_snap = cur_ops->start_gp_poll_exp();
while (!cur_ops->poll_gp_state_exp(gp_snap)) {
@ -1452,6 +1636,13 @@ rcu_torture_fakewriter(void *arg)
&rand);
}
break;
case RTWS_POLL_GET_EXP_FULL:
cur_ops->start_gp_poll_exp_full(&gp_snap_full);
while (!cur_ops->poll_gp_state_full(&gp_snap_full)) {
torture_hrtimeout_jiffies(torture_random(&rand) % 16,
&rand);
}
break;
case RTWS_SYNC:
cur_ops->sync();
break;
@ -1715,7 +1906,9 @@ rcutorture_loop_extend(int *readstate, struct torture_random_state *trsp,
*/
static bool rcu_torture_one_read(struct torture_random_state *trsp, long myid)
{
bool checkpolling = !(torture_random(trsp) & 0xfff);
unsigned long cookie;
struct rcu_gp_oldstate cookie_full;
int i;
unsigned long started;
unsigned long completed;
@ -1731,8 +1924,12 @@ static bool rcu_torture_one_read(struct torture_random_state *trsp, long myid)
WARN_ON_ONCE(!rcu_is_watching());
newstate = rcutorture_extend_mask(readstate, trsp);
rcutorture_one_extend(&readstate, newstate, trsp, rtrsp++);
if (checkpolling) {
if (cur_ops->get_gp_state && cur_ops->poll_gp_state)
cookie = cur_ops->get_gp_state();
if (cur_ops->get_gp_state_full && cur_ops->poll_gp_state_full)
cur_ops->get_gp_state_full(&cookie_full);
}
started = cur_ops->get_gp_seq();
ts = rcu_trace_clock_local();
p = rcu_dereference_check(rcu_torture_current,
@ -1766,6 +1963,7 @@ static bool rcu_torture_one_read(struct torture_random_state *trsp, long myid)
}
__this_cpu_inc(rcu_torture_batch[completed]);
preempt_enable();
if (checkpolling) {
if (cur_ops->get_gp_state && cur_ops->poll_gp_state)
WARN_ONCE(cur_ops->poll_gp_state(cookie),
"%s: Cookie check 2 failed %s(%d) %lu->%lu\n",
@ -1773,6 +1971,14 @@ static bool rcu_torture_one_read(struct torture_random_state *trsp, long myid)
rcu_torture_writer_state_getname(),
rcu_torture_writer_state,
cookie, cur_ops->get_gp_state());
if (cur_ops->get_gp_state_full && cur_ops->poll_gp_state_full)
WARN_ONCE(cur_ops->poll_gp_state_full(&cookie_full),
"%s: Cookie check 6 failed %s(%d) online %*pbl\n",
__func__,
rcu_torture_writer_state_getname(),
rcu_torture_writer_state,
cpumask_pr_args(cpu_online_mask));
}
rcutorture_one_extend(&readstate, 0, trsp, rtrsp);
WARN_ON_ONCE(readstate);
// This next splat is expected behavior if leakpointer, especially
@ -2600,12 +2806,12 @@ static int rcutorture_oom_notify(struct notifier_block *self,
for (i = 0; i < fwd_progress; i++)
ncbs += rcu_torture_fwd_prog_cbfree(&rfp[i]);
pr_info("%s: Freed %lu RCU callbacks.\n", __func__, ncbs);
rcu_barrier();
cur_ops->cb_barrier();
ncbs = 0;
for (i = 0; i < fwd_progress; i++)
ncbs += rcu_torture_fwd_prog_cbfree(&rfp[i]);
pr_info("%s: Freed %lu RCU callbacks.\n", __func__, ncbs);
rcu_barrier();
cur_ops->cb_barrier();
ncbs = 0;
for (i = 0; i < fwd_progress; i++)
ncbs += rcu_torture_fwd_prog_cbfree(&rfp[i]);

14
kernel/rcu/srcutiny.c
Просмотреть файл

@ -117,7 +117,7 @@ void srcu_drive_gp(struct work_struct *wp)
struct srcu_struct *ssp;
ssp = container_of(wp, struct srcu_struct, srcu_work);
if (ssp->srcu_gp_running || USHORT_CMP_GE(ssp->srcu_idx, READ_ONCE(ssp->srcu_idx_max)))
if (ssp->srcu_gp_running || ULONG_CMP_GE(ssp->srcu_idx, READ_ONCE(ssp->srcu_idx_max)))
return; /* Already running or nothing to do. */
/* Remove recently arrived callbacks and wait for readers. */
@ -150,17 +150,17 @@ void srcu_drive_gp(struct work_struct *wp)
* straighten that out.
*/
WRITE_ONCE(ssp->srcu_gp_running, false);
if (USHORT_CMP_LT(ssp->srcu_idx, READ_ONCE(ssp->srcu_idx_max)))
if (ULONG_CMP_LT(ssp->srcu_idx, READ_ONCE(ssp->srcu_idx_max)))
schedule_work(&ssp->srcu_work);
}
EXPORT_SYMBOL_GPL(srcu_drive_gp);
static void srcu_gp_start_if_needed(struct srcu_struct *ssp)
{
unsigned short cookie;
unsigned long cookie;
cookie = get_state_synchronize_srcu(ssp);
if (USHORT_CMP_GE(READ_ONCE(ssp->srcu_idx_max), cookie))
if (ULONG_CMP_GE(READ_ONCE(ssp->srcu_idx_max), cookie))
return;
WRITE_ONCE(ssp->srcu_idx_max, cookie);
if (!READ_ONCE(ssp->srcu_gp_running)) {
@ -215,7 +215,7 @@ unsigned long get_state_synchronize_srcu(struct srcu_struct *ssp)
barrier();
ret = (READ_ONCE(ssp->srcu_idx) + 3) & ~0x1;
barrier();
return ret & USHRT_MAX;
return ret;
}
EXPORT_SYMBOL_GPL(get_state_synchronize_srcu);
@ -240,10 +240,10 @@ EXPORT_SYMBOL_GPL(start_poll_synchronize_srcu);
*/
bool poll_state_synchronize_srcu(struct srcu_struct *ssp, unsigned long cookie)
{
bool ret = USHORT_CMP_GE(READ_ONCE(ssp->srcu_idx), cookie);
unsigned long cur_s = READ_ONCE(ssp->srcu_idx);
barrier();
return ret;
return ULONG_CMP_GE(cur_s, cookie) || ULONG_CMP_LT(cur_s, cookie - 3);
}
EXPORT_SYMBOL_GPL(poll_state_synchronize_srcu);

5
kernel/rcu/tasks.h
Просмотреть файл

@ -560,7 +560,7 @@ static int __noreturn rcu_tasks_kthread(void *arg)
static void synchronize_rcu_tasks_generic(struct rcu_tasks *rtp)
{
/* Complain if the scheduler has not started. */
RCU_LOCKDEP_WARN(rcu_scheduler_active == RCU_SCHEDULER_INACTIVE,
WARN_ONCE(rcu_scheduler_active == RCU_SCHEDULER_INACTIVE,
"synchronize_rcu_tasks called too soon");
// If the grace-period kthread is running, use it.
@ -1500,6 +1500,7 @@ static void rcu_tasks_trace_pregp_step(struct list_head *hop)
if (rcu_tasks_trace_pertask_prep(t, true))
trc_add_holdout(t, hop);
rcu_read_unlock();
cond_resched_tasks_rcu_qs();
}
// Only after all running tasks have been accounted for is it
@ -1520,6 +1521,7 @@ static void rcu_tasks_trace_pregp_step(struct list_head *hop)
raw_spin_lock_irqsave_rcu_node(rtpcp, flags);
}
raw_spin_unlock_irqrestore_rcu_node(rtpcp, flags);
cond_resched_tasks_rcu_qs();
}
// Re-enable CPU hotplug now that the holdout list is populated.
@ -1619,6 +1621,7 @@ static void check_all_holdout_tasks_trace(struct list_head *hop,
trc_del_holdout(t);
else if (needreport)
show_stalled_task_trace(t, firstreport);
cond_resched_tasks_rcu_qs();
}
// Re-enable CPU hotplug now that the holdout list scan has completed.

27
kernel/rcu/tiny.c
Просмотреть файл

@ -158,6 +158,10 @@ void synchronize_rcu(void)
}
EXPORT_SYMBOL_GPL(synchronize_rcu);
static void tiny_rcu_leak_callback(struct rcu_head *rhp)
{
}
/*
* Post an RCU callback to be invoked after the end of an RCU grace
* period. But since we have but one CPU, that would be after any
@ -165,9 +169,20 @@ EXPORT_SYMBOL_GPL(synchronize_rcu);
*/
void call_rcu(struct rcu_head *head, rcu_callback_t func)
{
static atomic_t doublefrees;
unsigned long flags;
debug_rcu_head_queue(head);
if (debug_rcu_head_queue(head)) {
if (atomic_inc_return(&doublefrees) < 4) {
pr_err("%s(): Double-freed CB %p->%pS()!!! ", __func__, head, head->func);
mem_dump_obj(head);
}
if (!__is_kvfree_rcu_offset((unsigned long)head->func))
WRITE_ONCE(head->func, tiny_rcu_leak_callback);
return;
}
head->func = func;
head->next = NULL;
@ -183,6 +198,16 @@ void call_rcu(struct rcu_head *head, rcu_callback_t func)
}
EXPORT_SYMBOL_GPL(call_rcu);
/*
* Store a grace-period-counter "cookie". For more information,
* see the Tree RCU header comment.
*/
void get_completed_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp)
{
rgosp->rgos_norm = RCU_GET_STATE_COMPLETED;
}
EXPORT_SYMBOL_GPL(get_completed_synchronize_rcu_full);
/*
* Return a grace-period-counter "cookie". For more information,
* see the Tree RCU header comment.

312
kernel/rcu/tree.c
Просмотреть файл

@ -76,6 +76,7 @@
/* Data structures. */
static DEFINE_PER_CPU_SHARED_ALIGNED(struct rcu_data, rcu_data) = {
.gpwrap = true,
#ifdef CONFIG_RCU_NOCB_CPU
.cblist.flags = SEGCBLIST_RCU_CORE,
#endif
@ -1755,6 +1756,8 @@ static noinline void rcu_gp_cleanup(void)
dump_blkd_tasks(rnp, 10);
WARN_ON_ONCE(rnp->qsmask);
WRITE_ONCE(rnp->gp_seq, new_gp_seq);
if (!rnp->parent)
smp_mb(); // Order against failing poll_state_synchronize_rcu_full().
rdp = this_cpu_ptr(&rcu_data);
if (rnp == rdp->mynode)
needgp = __note_gp_changes(rnp, rdp) || needgp;
@ -2341,8 +2344,8 @@ void rcu_sched_clock_irq(int user)
rcu_flavor_sched_clock_irq(user);
if (rcu_pending(user))
invoke_rcu_core();
if (user)
rcu_tasks_classic_qs(current, false);
if (user || rcu_is_cpu_rrupt_from_idle())
rcu_note_voluntary_context_switch(current);
lockdep_assert_irqs_disabled();
trace_rcu_utilization(TPS("End scheduler-tick"));
@ -2832,7 +2835,7 @@ EXPORT_SYMBOL_GPL(call_rcu);
/* Maximum number of jiffies to wait before draining a batch. */
#define KFREE_DRAIN_JIFFIES (HZ / 50)
#define KFREE_DRAIN_JIFFIES (5 * HZ)
#define KFREE_N_BATCHES 2
#define FREE_N_CHANNELS 2
@ -3093,6 +3096,21 @@ need_offload_krc(struct kfree_rcu_cpu *krcp)
return !!krcp->head;
}
static void
schedule_delayed_monitor_work(struct kfree_rcu_cpu *krcp)
{
long delay, delay_left;
delay = READ_ONCE(krcp->count) >= KVFREE_BULK_MAX_ENTR ? 1:KFREE_DRAIN_JIFFIES;
if (delayed_work_pending(&krcp->monitor_work)) {
delay_left = krcp->monitor_work.timer.expires - jiffies;
if (delay < delay_left)
mod_delayed_work(system_wq, &krcp->monitor_work, delay);
return;
}
queue_delayed_work(system_wq, &krcp->monitor_work, delay);
}
/*
* This function is invoked after the KFREE_DRAIN_JIFFIES timeout.
*/
@ -3150,7 +3168,7 @@ static void kfree_rcu_monitor(struct work_struct *work)
// work to repeat an attempt. Because previous batches are
// still in progress.
if (need_offload_krc(krcp))
schedule_delayed_work(&krcp->monitor_work, KFREE_DRAIN_JIFFIES);
schedule_delayed_monitor_work(krcp);
raw_spin_unlock_irqrestore(&krcp->lock, flags);
}
@ -3183,7 +3201,9 @@ static void fill_page_cache_func(struct work_struct *work)
bnode = (struct kvfree_rcu_bulk_data *)
__get_free_page(GFP_KERNEL | __GFP_NORETRY | __GFP_NOMEMALLOC | __GFP_NOWARN);
if (bnode) {
if (!bnode)
break;
raw_spin_lock_irqsave(&krcp->lock, flags);
pushed = put_cached_bnode(krcp, bnode);
raw_spin_unlock_irqrestore(&krcp->lock, flags);
@ -3193,7 +3213,6 @@ static void fill_page_cache_func(struct work_struct *work)
break;
}
}
}
atomic_set(&krcp->work_in_progress, 0);
atomic_set(&krcp->backoff_page_cache_fill, 0);
@ -3338,7 +3357,7 @@ void kvfree_call_rcu(struct rcu_head *head, rcu_callback_t func)
// Set timer to drain after KFREE_DRAIN_JIFFIES.
if (rcu_scheduler_active == RCU_SCHEDULER_RUNNING)
schedule_delayed_work(&krcp->monitor_work, KFREE_DRAIN_JIFFIES);
schedule_delayed_monitor_work(krcp);
unlock_return:
krc_this_cpu_unlock(krcp, flags);
@ -3371,7 +3390,7 @@ kfree_rcu_shrink_count(struct shrinker *shrink, struct shrink_control *sc)
atomic_set(&krcp->backoff_page_cache_fill, 1);
}
return count;
return count == 0 ? SHRINK_EMPTY : count;
}
static unsigned long
@ -3414,49 +3433,27 @@ void __init kfree_rcu_scheduler_running(void)
raw_spin_lock_irqsave(&krcp->lock, flags);
if (need_offload_krc(krcp))
schedule_delayed_work_on(cpu, &krcp->monitor_work, KFREE_DRAIN_JIFFIES);
schedule_delayed_monitor_work(krcp);
raw_spin_unlock_irqrestore(&krcp->lock, flags);
}
}
/*
* During early boot, any blocking grace-period wait automatically
* implies a grace period. Later on, this is never the case for PREEMPTION.
* implies a grace period.
*
* However, because a context switch is a grace period for !PREEMPTION, any
* blocking grace-period wait automatically implies a grace period if
* there is only one CPU online at any point time during execution of
* either synchronize_rcu() or synchronize_rcu_expedited(). It is OK to
* occasionally incorrectly indicate that there are multiple CPUs online
* when there was in fact only one the whole time, as this just adds some
* overhead: RCU still operates correctly.
* Later on, this could in theory be the case for kernels built with
* CONFIG_SMP=y && CONFIG_PREEMPTION=y running on a single CPU, but this
* is not a common case. Furthermore, this optimization would cause
* the rcu_gp_oldstate structure to expand by 50%, so this potential
* grace-period optimization is ignored once the scheduler is running.
*/
static int rcu_blocking_is_gp(void)
{
int ret;
// Invoking preempt_model_*() too early gets a splat.
if (rcu_scheduler_active == RCU_SCHEDULER_INACTIVE ||
preempt_model_full() || preempt_model_rt())
return rcu_scheduler_active == RCU_SCHEDULER_INACTIVE;
if (rcu_scheduler_active != RCU_SCHEDULER_INACTIVE)
return false;
might_sleep(); /* Check for RCU read-side critical section. */
preempt_disable();
/*
* If the rcu_state.n_online_cpus counter is equal to one,
* there is only one CPU, and that CPU sees all prior accesses
* made by any CPU that was online at the time of its access.
* Furthermore, if this counter is equal to one, its value cannot
* change until after the preempt_enable() below.
*
* Furthermore, if rcu_state.n_online_cpus is equal to one here,
* all later CPUs (both this one and any that come online later
* on) are guaranteed to see all accesses prior to this point
* in the code, without the need for additional memory barriers.
* Those memory barriers are provided by CPU-hotplug code.
*/
ret = READ_ONCE(rcu_state.n_online_cpus) <= 1;
preempt_enable();
return ret;
return true;
}
/**
@ -3499,29 +3496,58 @@ static int rcu_blocking_is_gp(void)
*/
void synchronize_rcu(void)
{
unsigned long flags;
struct rcu_node *rnp;
RCU_LOCKDEP_WARN(lock_is_held(&rcu_bh_lock_map) ||
lock_is_held(&rcu_lock_map) ||
lock_is_held(&rcu_sched_lock_map),
"Illegal synchronize_rcu() in RCU read-side critical section");
if (rcu_blocking_is_gp()) {
// Note well that this code runs with !PREEMPT && !SMP.
// In addition, all code that advances grace periods runs at
// process level. Therefore, this normal GP overlaps with
// other normal GPs only by being fully nested within them,
// which allows reuse of ->gp_seq_polled_snap.
rcu_poll_gp_seq_start_unlocked(&rcu_state.gp_seq_polled_snap);
rcu_poll_gp_seq_end_unlocked(&rcu_state.gp_seq_polled_snap);
if (rcu_init_invoked())
cond_resched_tasks_rcu_qs();
return; // Context allows vacuous grace periods.
}
if (!rcu_blocking_is_gp()) {
if (rcu_gp_is_expedited())
synchronize_rcu_expedited();
else
wait_rcu_gp(call_rcu);
return;
}
// Context allows vacuous grace periods.
// Note well that this code runs with !PREEMPT && !SMP.
// In addition, all code that advances grace periods runs at
// process level. Therefore, this normal GP overlaps with other
// normal GPs only by being fully nested within them, which allows
// reuse of ->gp_seq_polled_snap.
rcu_poll_gp_seq_start_unlocked(&rcu_state.gp_seq_polled_snap);
rcu_poll_gp_seq_end_unlocked(&rcu_state.gp_seq_polled_snap);
// Update the normal grace-period counters to record
// this grace period, but only those used by the boot CPU.
// The rcu_scheduler_starting() will take care of the rest of
// these counters.
local_irq_save(flags);
WARN_ON_ONCE(num_online_cpus() > 1);
rcu_state.gp_seq += (1 << RCU_SEQ_CTR_SHIFT);
for (rnp = this_cpu_ptr(&rcu_data)->mynode; rnp; rnp = rnp->parent)
rnp->gp_seq_needed = rnp->gp_seq = rcu_state.gp_seq;
local_irq_restore(flags);
}
EXPORT_SYMBOL_GPL(synchronize_rcu);
/**
* get_completed_synchronize_rcu_full - Return a full pre-completed polled state cookie
* @rgosp: Place to put state cookie
*
* Stores into @rgosp a value that will always be treated by functions
* like poll_state_synchronize_rcu_full() as a cookie whose grace period
* has already completed.
*/
void get_completed_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp)
{
rgosp->rgos_norm = RCU_GET_STATE_COMPLETED;
rgosp->rgos_exp = RCU_GET_STATE_COMPLETED;
}
EXPORT_SYMBOL_GPL(get_completed_synchronize_rcu_full);
/**
* get_state_synchronize_rcu - Snapshot current RCU state
*
@ -3541,21 +3567,42 @@ unsigned long get_state_synchronize_rcu(void)
EXPORT_SYMBOL_GPL(get_state_synchronize_rcu);
/**
* start_poll_synchronize_rcu - Snapshot and start RCU grace period
* get_state_synchronize_rcu_full - Snapshot RCU state, both normal and expedited
* @rgosp: location to place combined normal/expedited grace-period state
*
* Returns a cookie that is used by a later call to cond_synchronize_rcu()
* or poll_state_synchronize_rcu() to determine whether or not a full
* grace period has elapsed in the meantime. If the needed grace period
* is not already slated to start, notifies RCU core of the need for that
* grace period.
* Places the normal and expedited grace-period states in @rgosp. This
* state value can be passed to a later call to cond_synchronize_rcu_full()
* or poll_state_synchronize_rcu_full() to determine whether or not a
* grace period (whether normal or expedited) has elapsed in the meantime.
* The rcu_gp_oldstate structure takes up twice the memory of an unsigned
* long, but is guaranteed to see all grace periods. In contrast, the
* combined state occupies less memory, but can sometimes fail to take
* grace periods into account.
*
* Interrupts must be enabled for the case where it is necessary to awaken
* the grace-period kthread.
* This does not guarantee that the needed grace period will actually
* start.
*/
unsigned long start_poll_synchronize_rcu(void)
void get_state_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp)
{
struct rcu_node *rnp = rcu_get_root();
/*
* Any prior manipulation of RCU-protected data must happen
* before the loads from ->gp_seq and ->expedited_sequence.
*/
smp_mb(); /* ^^^ */
rgosp->rgos_norm = rcu_seq_snap(&rnp->gp_seq);
rgosp->rgos_exp = rcu_seq_snap(&rcu_state.expedited_sequence);
}
EXPORT_SYMBOL_GPL(get_state_synchronize_rcu_full);
/*
* Helper function for start_poll_synchronize_rcu() and
* start_poll_synchronize_rcu_full().
*/
static void start_poll_synchronize_rcu_common(void)
{
unsigned long flags;
unsigned long gp_seq = get_state_synchronize_rcu();
bool needwake;
struct rcu_data *rdp;
struct rcu_node *rnp;
@ -3575,17 +3622,57 @@ unsigned long start_poll_synchronize_rcu(void)
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
if (needwake)
rcu_gp_kthread_wake();
}
/**
* start_poll_synchronize_rcu - Snapshot and start RCU grace period
*
* Returns a cookie that is used by a later call to cond_synchronize_rcu()
* or poll_state_synchronize_rcu() to determine whether or not a full
* grace period has elapsed in the meantime. If the needed grace period
* is not already slated to start, notifies RCU core of the need for that
* grace period.
*
* Interrupts must be enabled for the case where it is necessary to awaken
* the grace-period kthread.
*/
unsigned long start_poll_synchronize_rcu(void)
{
unsigned long gp_seq = get_state_synchronize_rcu();
start_poll_synchronize_rcu_common();
return gp_seq;
}
EXPORT_SYMBOL_GPL(start_poll_synchronize_rcu);
/**
* poll_state_synchronize_rcu - Conditionally wait for an RCU grace period
* start_poll_synchronize_rcu_full - Take a full snapshot and start RCU grace period
* @rgosp: value from get_state_synchronize_rcu_full() or start_poll_synchronize_rcu_full()
*
* Places the normal and expedited grace-period states in *@rgos. This
* state value can be passed to a later call to cond_synchronize_rcu_full()
* or poll_state_synchronize_rcu_full() to determine whether or not a
* grace period (whether normal or expedited) has elapsed in the meantime.
* If the needed grace period is not already slated to start, notifies
* RCU core of the need for that grace period.
*
* Interrupts must be enabled for the case where it is necessary to awaken
* the grace-period kthread.
*/
void start_poll_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp)
{
get_state_synchronize_rcu_full(rgosp);
start_poll_synchronize_rcu_common();
}
EXPORT_SYMBOL_GPL(start_poll_synchronize_rcu_full);
/**
* poll_state_synchronize_rcu - Has the specified RCU grace period completed?
* @oldstate: value from get_state_synchronize_rcu() or start_poll_synchronize_rcu()
*
* If a full RCU grace period has elapsed since the earlier call from
* which oldstate was obtained, return @true, otherwise return @false.
* which @oldstate was obtained, return @true, otherwise return @false.
* If @false is returned, it is the caller's responsibility to invoke this
* function later on until it does return @true. Alternatively, the caller
* can explicitly wait for a grace period, for example, by passing @oldstate
@ -3594,10 +3681,11 @@ EXPORT_SYMBOL_GPL(start_poll_synchronize_rcu);
* Yes, this function does not take counter wrap into account.
* But counter wrap is harmless. If the counter wraps, we have waited for
* more than a billion grace periods (and way more on a 64-bit system!).
* Those needing to keep oldstate values for very long time periods
* (many hours even on 32-bit systems) should check them occasionally
* and either refresh them or set a flag indicating that the grace period
* has completed.
* Those needing to keep old state values for very long time periods
* (many hours even on 32-bit systems) should check them occasionally and
* either refresh them or set a flag indicating that the grace period has
* completed. Alternatively, they can use get_completed_synchronize_rcu()
* to get a guaranteed-completed grace-period state.
*
* This function provides the same memory-ordering guarantees that
* would be provided by a synchronize_rcu() that was invoked at the call
@ -3616,8 +3704,56 @@ bool poll_state_synchronize_rcu(unsigned long oldstate)
EXPORT_SYMBOL_GPL(poll_state_synchronize_rcu);
/**
* cond_synchronize_rcu - Conditionally wait for an RCU grace period
* poll_state_synchronize_rcu_full - Has the specified RCU grace period completed?
* @rgosp: value from get_state_synchronize_rcu_full() or start_poll_synchronize_rcu_full()
*
* If a full RCU grace period has elapsed since the earlier call from
* which *rgosp was obtained, return @true, otherwise return @false.
* If @false is returned, it is the caller's responsibility to invoke this
* function later on until it does return @true. Alternatively, the caller
* can explicitly wait for a grace period, for example, by passing @rgosp
* to cond_synchronize_rcu() or by directly invoking synchronize_rcu().
*
* Yes, this function does not take counter wrap into account.
* But counter wrap is harmless. If the counter wraps, we have waited
* for more than a billion grace periods (and way more on a 64-bit
* system!). Those needing to keep rcu_gp_oldstate values for very
* long time periods (many hours even on 32-bit systems) should check
* them occasionally and either refresh them or set a flag indicating
* that the grace period has completed. Alternatively, they can use
* get_completed_synchronize_rcu_full() to get a guaranteed-completed
* grace-period state.
*
* This function provides the same memory-ordering guarantees that would
* be provided by a synchronize_rcu() that was invoked at the call to
* the function that provided @rgosp, and that returned at the end of this
* function. And this guarantee requires that the root rcu_node structure's
* ->gp_seq field be checked instead of that of the rcu_state structure.
* The problem is that the just-ending grace-period's callbacks can be
* invoked between the time that the root rcu_node structure's ->gp_seq
* field is updated and the time that the rcu_state structure's ->gp_seq
* field is updated. Therefore, if a single synchronize_rcu() is to
* cause a subsequent poll_state_synchronize_rcu_full() to return @true,
* then the root rcu_node structure is the one that needs to be polled.
*/
bool poll_state_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp)
{
struct rcu_node *rnp = rcu_get_root();
smp_mb(); // Order against root rcu_node structure grace-period cleanup.
if (rgosp->rgos_norm == RCU_GET_STATE_COMPLETED ||
rcu_seq_done_exact(&rnp->gp_seq, rgosp->rgos_norm) ||
rgosp->rgos_exp == RCU_GET_STATE_COMPLETED ||
rcu_seq_done_exact(&rcu_state.expedited_sequence, rgosp->rgos_exp)) {
smp_mb(); /* Ensure GP ends before subsequent accesses. */
return true;
}
return false;
}
EXPORT_SYMBOL_GPL(poll_state_synchronize_rcu_full);
/**
* cond_synchronize_rcu - Conditionally wait for an RCU grace period
* @oldstate: value from get_state_synchronize_rcu(), start_poll_synchronize_rcu(), or start_poll_synchronize_rcu_expedited()
*
* If a full RCU grace period has elapsed since the earlier call to
@ -3641,6 +3777,33 @@ void cond_synchronize_rcu(unsigned long oldstate)
}
EXPORT_SYMBOL_GPL(cond_synchronize_rcu);
/**
* cond_synchronize_rcu_full - Conditionally wait for an RCU grace period
* @rgosp: value from get_state_synchronize_rcu_full(), start_poll_synchronize_rcu_full(), or start_poll_synchronize_rcu_expedited_full()
*
* If a full RCU grace period has elapsed since the call to
* get_state_synchronize_rcu_full(), start_poll_synchronize_rcu_full(),
* or start_poll_synchronize_rcu_expedited_full() from which @rgosp was
* obtained, just return. Otherwise, invoke synchronize_rcu() to wait
* for a full grace period.
*
* Yes, this function does not take counter wrap into account.
* But counter wrap is harmless. If the counter wraps, we have waited for
* more than 2 billion grace periods (and way more on a 64-bit system!),
* so waiting for a couple of additional grace periods should be just fine.
*
* This function provides the same memory-ordering guarantees that
* would be provided by a synchronize_rcu() that was invoked at the call
* to the function that provided @rgosp and that returned at the end of
* this function.
*/
void cond_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp)
{
if (!poll_state_synchronize_rcu_full(rgosp))
synchronize_rcu();
}
EXPORT_SYMBOL_GPL(cond_synchronize_rcu_full);
/*
* Check to see if there is any immediate RCU-related work to be done by
* the current CPU, returning 1 if so and zero otherwise. The checks are
@ -4312,9 +4475,20 @@ early_initcall(rcu_spawn_gp_kthread);
*/
void rcu_scheduler_starting(void)
{
unsigned long flags;
struct rcu_node *rnp;
WARN_ON(num_online_cpus() != 1);
WARN_ON(nr_context_switches() > 0);
rcu_test_sync_prims();
// Fix up the ->gp_seq counters.
local_irq_save(flags);
rcu_for_each_node_breadth_first(rnp)
rnp->gp_seq_needed = rnp->gp_seq = rcu_state.gp_seq;
local_irq_restore(flags);
// Switch out of early boot mode.
rcu_scheduler_active = RCU_SCHEDULER_INIT;
rcu_test_sync_prims();
}

57
kernel/rcu/tree_exp.h
Просмотреть файл

@ -828,11 +828,13 @@ static void rcu_exp_handler(void *unused)
{
struct rcu_data *rdp = this_cpu_ptr(&rcu_data);
struct rcu_node *rnp = rdp->mynode;
bool preempt_bh_enabled = !(preempt_count() & (PREEMPT_MASK | SOFTIRQ_MASK));
if (!(READ_ONCE(rnp->expmask) & rdp->grpmask) ||
__this_cpu_read(rcu_data.cpu_no_qs.b.exp))
return;
if (rcu_is_cpu_rrupt_from_idle()) {
if (rcu_is_cpu_rrupt_from_idle() ||
(IS_ENABLED(CONFIG_PREEMPT_COUNT) && preempt_bh_enabled)) {
rcu_report_exp_rdp(this_cpu_ptr(&rcu_data));
return;
}
@ -906,6 +908,7 @@ static int rcu_print_task_exp_stall(struct rcu_node *rnp)
void synchronize_rcu_expedited(void)
{
bool boottime = (rcu_scheduler_active == RCU_SCHEDULER_INIT);
unsigned long flags;
struct rcu_exp_work rew;
struct rcu_node *rnp;
unsigned long s;
@ -924,8 +927,11 @@ void synchronize_rcu_expedited(void)
// them, which allows reuse of ->gp_seq_polled_exp_snap.
rcu_poll_gp_seq_start_unlocked(&rcu_state.gp_seq_polled_exp_snap);
rcu_poll_gp_seq_end_unlocked(&rcu_state.gp_seq_polled_exp_snap);
if (rcu_init_invoked())
cond_resched();
local_irq_save(flags);
WARN_ON_ONCE(num_online_cpus() > 1);
rcu_state.expedited_sequence += (1 << RCU_SEQ_CTR_SHIFT);
local_irq_restore(flags);
return; // Context allows vacuous grace periods.
}
@ -1027,6 +1033,24 @@ unsigned long start_poll_synchronize_rcu_expedited(void)
}
EXPORT_SYMBOL_GPL(start_poll_synchronize_rcu_expedited);
/**
* start_poll_synchronize_rcu_expedited_full - Take a full snapshot and start expedited grace period
* @rgosp: Place to put snapshot of grace-period state
*
* Places the normal and expedited grace-period states in rgosp. This
* state value can be passed to a later call to cond_synchronize_rcu_full()
* or poll_state_synchronize_rcu_full() to determine whether or not a
* grace period (whether normal or expedited) has elapsed in the meantime.
* If the needed expedited grace period is not already slated to start,
* initiates that grace period.
*/
void start_poll_synchronize_rcu_expedited_full(struct rcu_gp_oldstate *rgosp)
{
get_state_synchronize_rcu_full(rgosp);
(void)start_poll_synchronize_rcu_expedited();
}
EXPORT_SYMBOL_GPL(start_poll_synchronize_rcu_expedited_full);
/**
* cond_synchronize_rcu_expedited - Conditionally wait for an expedited RCU grace period
*
@ -1053,3 +1077,30 @@ void cond_synchronize_rcu_expedited(unsigned long oldstate)
synchronize_rcu_expedited();
}
EXPORT_SYMBOL_GPL(cond_synchronize_rcu_expedited);
/**
* cond_synchronize_rcu_expedited_full - Conditionally wait for an expedited RCU grace period
* @rgosp: value from get_state_synchronize_rcu_full(), start_poll_synchronize_rcu_full(), or start_poll_synchronize_rcu_expedited_full()
*
* If a full RCU grace period has elapsed since the call to
* get_state_synchronize_rcu_full(), start_poll_synchronize_rcu_full(),
* or start_poll_synchronize_rcu_expedited_full() from which @rgosp was
* obtained, just return. Otherwise, invoke synchronize_rcu_expedited()
* to wait for a full grace period.
*
* Yes, this function does not take counter wrap into account.
* But counter wrap is harmless. If the counter wraps, we have waited for
* more than 2 billion grace periods (and way more on a 64-bit system!),
* so waiting for a couple of additional grace periods should be just fine.
*
* This function provides the same memory-ordering guarantees that
* would be provided by a synchronize_rcu() that was invoked at the call
* to the function that provided @rgosp and that returned at the end of
* this function.
*/
void cond_synchronize_rcu_expedited_full(struct rcu_gp_oldstate *rgosp)
{
if (!poll_state_synchronize_rcu_full(rgosp))
synchronize_rcu_expedited();
}
EXPORT_SYMBOL_GPL(cond_synchronize_rcu_expedited_full);

10
kernel/rcu/tree_nocb.h
Просмотреть файл

@ -1111,7 +1111,7 @@ int rcu_nocb_cpu_deoffload(int cpu)
if (!ret)
cpumask_clear_cpu(cpu, rcu_nocb_mask);
} else {
pr_info("NOCB: Can't CB-deoffload an offline CPU\n");
pr_info("NOCB: Cannot CB-deoffload offline CPU %d\n", rdp->cpu);
ret = -EINVAL;
}
}
@ -1196,7 +1196,7 @@ int rcu_nocb_cpu_offload(int cpu)
if (!ret)
cpumask_set_cpu(cpu, rcu_nocb_mask);
} else {
pr_info("NOCB: Can't CB-offload an offline CPU\n");
pr_info("NOCB: Cannot CB-offload offline CPU %d\n", rdp->cpu);
ret = -EINVAL;
}
}
@ -1452,8 +1452,8 @@ static void show_rcu_nocb_gp_state(struct rcu_data *rdp)
(long)rdp->nocb_gp_seq,
rnp->grplo, rnp->grphi, READ_ONCE(rdp->nocb_gp_loops),
rdp->nocb_gp_kthread ? task_state_to_char(rdp->nocb_gp_kthread) : '.',
rdp->nocb_cb_kthread ? (int)task_cpu(rdp->nocb_gp_kthread) : -1,
show_rcu_should_be_on_cpu(rdp->nocb_cb_kthread));
rdp->nocb_gp_kthread ? (int)task_cpu(rdp->nocb_gp_kthread) : -1,
show_rcu_should_be_on_cpu(rdp->nocb_gp_kthread));
}
/* Dump out nocb kthread state for the specified rcu_data structure. */
@ -1497,7 +1497,7 @@ static void show_rcu_nocb_state(struct rcu_data *rdp)
".B"[!!rcu_cblist_n_cbs(&rdp->nocb_bypass)],
rcu_segcblist_n_cbs(&rdp->cblist),
rdp->nocb_cb_kthread ? task_state_to_char(rdp->nocb_cb_kthread) : '.',
rdp->nocb_cb_kthread ? (int)task_cpu(rdp->nocb_gp_kthread) : -1,
rdp->nocb_cb_kthread ? (int)task_cpu(rdp->nocb_cb_kthread) : -1,
show_rcu_should_be_on_cpu(rdp->nocb_cb_kthread));
/* It is OK for GP kthreads to have GP state. */

26
kernel/rcu/tree_plugin.h
Просмотреть файл

@ -641,7 +641,8 @@ static void rcu_read_unlock_special(struct task_struct *t)
expboost = (t->rcu_blocked_node && READ_ONCE(t->rcu_blocked_node->exp_tasks)) ||
(rdp->grpmask & READ_ONCE(rnp->expmask)) ||
IS_ENABLED(CONFIG_RCU_STRICT_GRACE_PERIOD) ||
(IS_ENABLED(CONFIG_RCU_STRICT_GRACE_PERIOD) &&
((rdp->grpmask & READ_ONCE(rnp->qsmask)) || t->rcu_blocked_node)) ||
(IS_ENABLED(CONFIG_RCU_BOOST) && irqs_were_disabled &&
t->rcu_blocked_node);
// Need to defer quiescent state until everything is enabled.
@ -718,9 +719,6 @@ static void rcu_flavor_sched_clock_irq(int user)
struct task_struct *t = current;
lockdep_assert_irqs_disabled();
if (user || rcu_is_cpu_rrupt_from_idle()) {
rcu_note_voluntary_context_switch(current);
}
if (rcu_preempt_depth() > 0 ||
(preempt_count() & (PREEMPT_MASK | SOFTIRQ_MASK))) {
/* No QS, force context switch if deferred. */
@ -824,6 +822,7 @@ void rcu_read_unlock_strict(void)
if (irqs_disabled() || preempt_count() || !rcu_state.gp_kthread)
return;
rdp = this_cpu_ptr(&rcu_data);
rdp->cpu_no_qs.b.norm = false;
rcu_report_qs_rdp(rdp);
udelay(rcu_unlock_delay);
}
@ -869,7 +868,7 @@ void rcu_all_qs(void)
if (!raw_cpu_read(rcu_data.rcu_urgent_qs))
return;
preempt_disable();
preempt_disable(); // For CONFIG_PREEMPT_COUNT=y kernels
/* Load rcu_urgent_qs before other flags. */
if (!smp_load_acquire(this_cpu_ptr(&rcu_data.rcu_urgent_qs))) {
preempt_enable();
@ -931,10 +930,13 @@ static notrace bool rcu_preempt_need_deferred_qs(struct task_struct *t)
return false;
}
// Except that we do need to respond to a request by an expedited grace
// period for a quiescent state from this CPU. Note that requests from
// tasks are handled when removing the task from the blocked-tasks list
// below.
// Except that we do need to respond to a request by an expedited
// grace period for a quiescent state from this CPU. Note that in
// non-preemptible kernels, there can be no context switches within RCU
// read-side critical sections, which in turn means that the leaf rcu_node
// structure's blocked-tasks list is always empty. is therefore no need to
// actually check it. Instead, a quiescent state from this CPU suffices,
// and this function is only called from such a quiescent state.
notrace void rcu_preempt_deferred_qs(struct task_struct *t)
{
struct rcu_data *rdp = this_cpu_ptr(&rcu_data);
@ -972,7 +974,6 @@ static void rcu_flavor_sched_clock_irq(int user)
* neither access nor modify, at least not while the
* corresponding CPU is online.
*/
rcu_qs();
}
}
@ -1238,8 +1239,11 @@ static void rcu_boost_kthread_setaffinity(struct rcu_node *rnp, int outgoingcpu)
cpu != outgoingcpu)
cpumask_set_cpu(cpu, cm);
cpumask_and(cm, cm, housekeeping_cpumask(HK_TYPE_RCU));
if (cpumask_empty(cm))
if (cpumask_empty(cm)) {
cpumask_copy(cm, housekeeping_cpumask(HK_TYPE_RCU));
if (outgoingcpu >= 0)
cpumask_clear_cpu(outgoingcpu, cm);
}
set_cpus_allowed_ptr(t, cm);
mutex_unlock(&rnp->boost_kthread_mutex);
free_cpumask_var(cm);

3
kernel/rcu/tree_stall.h
Просмотреть файл

@ -368,7 +368,7 @@ static void rcu_dump_cpu_stacks(void)
if (rnp->qsmask & leaf_node_cpu_bit(rnp, cpu)) {
if (cpu_is_offline(cpu))
pr_err("Offline CPU %d blocking current GP.\n", cpu);
else if (!trigger_single_cpu_backtrace(cpu))
else
dump_cpu_task(cpu);
}
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
@ -511,7 +511,6 @@ static void rcu_check_gp_kthread_starvation(void)
pr_err("RCU GP kthread last ran on offline CPU %d.\n", cpu);
} else {
pr_err("Stack dump where RCU GP kthread last ran:\n");
if (!trigger_single_cpu_backtrace(cpu))
dump_cpu_task(cpu);
}
}

14
kernel/sched/core.c
Просмотреть файл

@ -73,6 +73,7 @@
#include <uapi/linux/sched/types.h>
#include <asm/irq_regs.h>
#include <asm/switch_to.h>
#include <asm/tlb.h>
@ -11183,6 +11184,19 @@ struct cgroup_subsys cpu_cgrp_subsys = {
void dump_cpu_task(int cpu)
{
if (cpu == smp_processor_id() && in_hardirq()) {
struct pt_regs *regs;
regs = get_irq_regs();
if (regs) {
show_regs(regs);
return;
}
}
if (trigger_single_cpu_backtrace(cpu))
return;
pr_info("Task dump for CPU %d:\n", cpu);
sched_show_task(cpu_curr(cpu));
}

1
kernel/smp.c
Просмотреть файл

@ -370,7 +370,6 @@ static bool csd_lock_wait_toolong(struct __call_single_data *csd, u64 ts0, u64 *
if (cpu >= 0) {
if (static_branch_unlikely(&csdlock_debug_extended))
csd_lock_print_extended(csd, cpu);
if (!trigger_single_cpu_backtrace(cpu))
dump_cpu_task(cpu);
if (!cpu_cur_csd) {
pr_alert("csd: Re-sending CSD lock (#%d) IPI from CPU#%02d to CPU#%02d\n", *bug_id, raw_smp_processor_id(), cpu);