linux/kernel/srcu.c

651 lines
20 KiB
C
Raw Normal View History

[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
/*
* Sleepable Read-Copy Update mechanism for mutual exclusion.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
*
* Copyright (C) IBM Corporation, 2006
*
* Author: Paul McKenney <paulmck@us.ibm.com>
*
* For detailed explanation of Read-Copy Update mechanism see -
* Documentation/RCU/ *.txt
*
*/
#include <linux/export.h>
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
#include <linux/mutex.h>
#include <linux/percpu.h>
#include <linux/preempt.h>
#include <linux/rcupdate.h>
#include <linux/sched.h>
#include <linux/smp.h>
#include <linux/delay.h>
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
#include <linux/srcu.h>
/*
* Initialize an rcu_batch structure to empty.
*/
static inline void rcu_batch_init(struct rcu_batch *b)
{
b->head = NULL;
b->tail = &b->head;
}
/*
* Enqueue a callback onto the tail of the specified rcu_batch structure.
*/
static inline void rcu_batch_queue(struct rcu_batch *b, struct rcu_head *head)
{
*b->tail = head;
b->tail = &head->next;
}
/*
* Is the specified rcu_batch structure empty?
*/
static inline bool rcu_batch_empty(struct rcu_batch *b)
{
return b->tail == &b->head;
}
/*
* Remove the callback at the head of the specified rcu_batch structure
* and return a pointer to it, or return NULL if the structure is empty.
*/
static inline struct rcu_head *rcu_batch_dequeue(struct rcu_batch *b)
{
struct rcu_head *head;
if (rcu_batch_empty(b))
return NULL;
head = b->head;
b->head = head->next;
if (b->tail == &head->next)
rcu_batch_init(b);
return head;
}
/*
* Move all callbacks from the rcu_batch structure specified by "from" to
* the structure specified by "to".
*/
static inline void rcu_batch_move(struct rcu_batch *to, struct rcu_batch *from)
{
if (!rcu_batch_empty(from)) {
*to->tail = from->head;
to->tail = from->tail;
rcu_batch_init(from);
}
}
/* single-thread state-machine */
static void process_srcu(struct work_struct *work);
rcu: Introduce lockdep-based checking to RCU read-side primitives Inspection is proving insufficient to catch all RCU misuses, which is understandable given that rcu_dereference() might be protected by any of four different flavors of RCU (RCU, RCU-bh, RCU-sched, and SRCU), and might also/instead be protected by any of a number of locking primitives. It is therefore time to enlist the aid of lockdep. This set of patches is inspired by earlier work by Peter Zijlstra and Thomas Gleixner, and takes the following approach: o Set up separate lockdep classes for RCU, RCU-bh, and RCU-sched. o Set up separate lockdep classes for each instance of SRCU. o Create primitives that check for being in an RCU read-side critical section. These return exact answers if lockdep is fully enabled, but if unsure, report being in an RCU read-side critical section. (We want to avoid false positives!) The primitives are: For RCU: rcu_read_lock_held(void) For RCU-bh: rcu_read_lock_bh_held(void) For RCU-sched: rcu_read_lock_sched_held(void) For SRCU: srcu_read_lock_held(struct srcu_struct *sp) o Add rcu_dereference_check(), which takes a second argument in which one places a boolean expression based on the above primitives and/or lockdep_is_held(). o A new kernel configuration parameter, CONFIG_PROVE_RCU, enables rcu_dereference_check(). This depends on CONFIG_PROVE_LOCKING, and should be quite helpful during the transition period while CONFIG_PROVE_RCU-unaware patches are in flight. The existing rcu_dereference() primitive does no checking, but upcoming patches will change that. Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1266887105-1528-1-git-send-email-paulmck@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-23 02:04:45 +01:00
static int init_srcu_struct_fields(struct srcu_struct *sp)
{
sp->completed = 0;
spin_lock_init(&sp->queue_lock);
sp->running = false;
rcu_batch_init(&sp->batch_queue);
rcu_batch_init(&sp->batch_check0);
rcu_batch_init(&sp->batch_check1);
rcu_batch_init(&sp->batch_done);
INIT_DELAYED_WORK(&sp->work, process_srcu);
rcu: Introduce lockdep-based checking to RCU read-side primitives Inspection is proving insufficient to catch all RCU misuses, which is understandable given that rcu_dereference() might be protected by any of four different flavors of RCU (RCU, RCU-bh, RCU-sched, and SRCU), and might also/instead be protected by any of a number of locking primitives. It is therefore time to enlist the aid of lockdep. This set of patches is inspired by earlier work by Peter Zijlstra and Thomas Gleixner, and takes the following approach: o Set up separate lockdep classes for RCU, RCU-bh, and RCU-sched. o Set up separate lockdep classes for each instance of SRCU. o Create primitives that check for being in an RCU read-side critical section. These return exact answers if lockdep is fully enabled, but if unsure, report being in an RCU read-side critical section. (We want to avoid false positives!) The primitives are: For RCU: rcu_read_lock_held(void) For RCU-bh: rcu_read_lock_bh_held(void) For RCU-sched: rcu_read_lock_sched_held(void) For SRCU: srcu_read_lock_held(struct srcu_struct *sp) o Add rcu_dereference_check(), which takes a second argument in which one places a boolean expression based on the above primitives and/or lockdep_is_held(). o A new kernel configuration parameter, CONFIG_PROVE_RCU, enables rcu_dereference_check(). This depends on CONFIG_PROVE_LOCKING, and should be quite helpful during the transition period while CONFIG_PROVE_RCU-unaware patches are in flight. The existing rcu_dereference() primitive does no checking, but upcoming patches will change that. Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1266887105-1528-1-git-send-email-paulmck@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-23 02:04:45 +01:00
sp->per_cpu_ref = alloc_percpu(struct srcu_struct_array);
return sp->per_cpu_ref ? 0 : -ENOMEM;
}
#ifdef CONFIG_DEBUG_LOCK_ALLOC
int __init_srcu_struct(struct srcu_struct *sp, const char *name,
struct lock_class_key *key)
{
/* Don't re-initialize a lock while it is held. */
debug_check_no_locks_freed((void *)sp, sizeof(*sp));
lockdep_init_map(&sp->dep_map, name, key, 0);
return init_srcu_struct_fields(sp);
}
EXPORT_SYMBOL_GPL(__init_srcu_struct);
#else /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
/**
* init_srcu_struct - initialize a sleep-RCU structure
* @sp: structure to initialize.
*
* Must invoke this on a given srcu_struct before passing that srcu_struct
* to any other function. Each srcu_struct represents a separate domain
* of SRCU protection.
*/
int init_srcu_struct(struct srcu_struct *sp)
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
{
rcu: Introduce lockdep-based checking to RCU read-side primitives Inspection is proving insufficient to catch all RCU misuses, which is understandable given that rcu_dereference() might be protected by any of four different flavors of RCU (RCU, RCU-bh, RCU-sched, and SRCU), and might also/instead be protected by any of a number of locking primitives. It is therefore time to enlist the aid of lockdep. This set of patches is inspired by earlier work by Peter Zijlstra and Thomas Gleixner, and takes the following approach: o Set up separate lockdep classes for RCU, RCU-bh, and RCU-sched. o Set up separate lockdep classes for each instance of SRCU. o Create primitives that check for being in an RCU read-side critical section. These return exact answers if lockdep is fully enabled, but if unsure, report being in an RCU read-side critical section. (We want to avoid false positives!) The primitives are: For RCU: rcu_read_lock_held(void) For RCU-bh: rcu_read_lock_bh_held(void) For RCU-sched: rcu_read_lock_sched_held(void) For SRCU: srcu_read_lock_held(struct srcu_struct *sp) o Add rcu_dereference_check(), which takes a second argument in which one places a boolean expression based on the above primitives and/or lockdep_is_held(). o A new kernel configuration parameter, CONFIG_PROVE_RCU, enables rcu_dereference_check(). This depends on CONFIG_PROVE_LOCKING, and should be quite helpful during the transition period while CONFIG_PROVE_RCU-unaware patches are in flight. The existing rcu_dereference() primitive does no checking, but upcoming patches will change that. Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1266887105-1528-1-git-send-email-paulmck@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-23 02:04:45 +01:00
return init_srcu_struct_fields(sp);
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
}
EXPORT_SYMBOL_GPL(init_srcu_struct);
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
rcu: Introduce lockdep-based checking to RCU read-side primitives Inspection is proving insufficient to catch all RCU misuses, which is understandable given that rcu_dereference() might be protected by any of four different flavors of RCU (RCU, RCU-bh, RCU-sched, and SRCU), and might also/instead be protected by any of a number of locking primitives. It is therefore time to enlist the aid of lockdep. This set of patches is inspired by earlier work by Peter Zijlstra and Thomas Gleixner, and takes the following approach: o Set up separate lockdep classes for RCU, RCU-bh, and RCU-sched. o Set up separate lockdep classes for each instance of SRCU. o Create primitives that check for being in an RCU read-side critical section. These return exact answers if lockdep is fully enabled, but if unsure, report being in an RCU read-side critical section. (We want to avoid false positives!) The primitives are: For RCU: rcu_read_lock_held(void) For RCU-bh: rcu_read_lock_bh_held(void) For RCU-sched: rcu_read_lock_sched_held(void) For SRCU: srcu_read_lock_held(struct srcu_struct *sp) o Add rcu_dereference_check(), which takes a second argument in which one places a boolean expression based on the above primitives and/or lockdep_is_held(). o A new kernel configuration parameter, CONFIG_PROVE_RCU, enables rcu_dereference_check(). This depends on CONFIG_PROVE_LOCKING, and should be quite helpful during the transition period while CONFIG_PROVE_RCU-unaware patches are in flight. The existing rcu_dereference() primitive does no checking, but upcoming patches will change that. Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1266887105-1528-1-git-send-email-paulmck@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-23 02:04:45 +01:00
#endif /* #else #ifdef CONFIG_DEBUG_LOCK_ALLOC */
/*
* Returns approximate total of the readers' ->seq[] values for the
* rank of per-CPU counters specified by idx.
*/
static unsigned long srcu_readers_seq_idx(struct srcu_struct *sp, int idx)
{
int cpu;
unsigned long sum = 0;
unsigned long t;
for_each_possible_cpu(cpu) {
t = ACCESS_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->seq[idx]);
sum += t;
}
return sum;
}
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
/*
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
* Returns approximate number of readers active on the specified rank
* of the per-CPU ->c[] counters.
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
*/
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
static unsigned long srcu_readers_active_idx(struct srcu_struct *sp, int idx)
{
int cpu;
unsigned long sum = 0;
unsigned long t;
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
for_each_possible_cpu(cpu) {
t = ACCESS_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->c[idx]);
sum += t;
}
return sum;
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
}
/*
* Return true if the number of pre-existing readers is determined to
* be stably zero. An example unstable zero can occur if the call
* to srcu_readers_active_idx() misses an __srcu_read_lock() increment,
* but due to task migration, sees the corresponding __srcu_read_unlock()
* decrement. This can happen because srcu_readers_active_idx() takes
* time to sum the array, and might in fact be interrupted or preempted
* partway through the summation.
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
*/
static bool srcu_readers_active_idx_check(struct srcu_struct *sp, int idx)
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
{
unsigned long seq;
seq = srcu_readers_seq_idx(sp, idx);
/*
* The following smp_mb() A pairs with the smp_mb() B located in
* __srcu_read_lock(). This pairing ensures that if an
* __srcu_read_lock() increments its counter after the summation
* in srcu_readers_active_idx(), then the corresponding SRCU read-side
* critical section will see any changes made prior to the start
* of the current SRCU grace period.
*
* Also, if the above call to srcu_readers_seq_idx() saw the
* increment of ->seq[], then the call to srcu_readers_active_idx()
* must see the increment of ->c[].
*/
smp_mb(); /* A */
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
/*
* Note that srcu_readers_active_idx() can incorrectly return
* zero even though there is a pre-existing reader throughout.
* To see this, suppose that task A is in a very long SRCU
* read-side critical section that started on CPU 0, and that
* no other reader exists, so that the sum of the counters
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
* is equal to one. Then suppose that task B starts executing
* srcu_readers_active_idx(), summing up to CPU 1, and then that
* task C starts reading on CPU 0, so that its increment is not
* summed, but finishes reading on CPU 2, so that its decrement
* -is- summed. Then when task B completes its sum, it will
* incorrectly get zero, despite the fact that task A has been
* in its SRCU read-side critical section the whole time.
*
* We therefore do a validation step should srcu_readers_active_idx()
* return zero.
*/
if (srcu_readers_active_idx(sp, idx) != 0)
return false;
/*
* The remainder of this function is the validation step.
* The following smp_mb() D pairs with the smp_mb() C in
* __srcu_read_unlock(). If the __srcu_read_unlock() was seen
* by srcu_readers_active_idx() above, then any destructive
* operation performed after the grace period will happen after
* the corresponding SRCU read-side critical section.
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
*
* Note that there can be at most NR_CPUS worth of readers using
* the old index, which is not enough to overflow even a 32-bit
* integer. (Yes, this does mean that systems having more than
* a billion or so CPUs need to be 64-bit systems.) Therefore,
* the sum of the ->seq[] counters cannot possibly overflow.
* Therefore, the only way that the return values of the two
* calls to srcu_readers_seq_idx() can be equal is if there were
* no increments of the corresponding rank of ->seq[] counts
* in the interim. But the missed-increment scenario laid out
* above includes an increment of the ->seq[] counter by
* the corresponding __srcu_read_lock(). Therefore, if this
* scenario occurs, the return values from the two calls to
* srcu_readers_seq_idx() will differ, and thus the validation
* step below suffices.
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
*/
smp_mb(); /* D */
return srcu_readers_seq_idx(sp, idx) == seq;
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
}
/**
* srcu_readers_active - returns approximate number of readers.
* @sp: which srcu_struct to count active readers (holding srcu_read_lock).
*
* Note that this is not an atomic primitive, and can therefore suffer
* severe errors when invoked on an active srcu_struct. That said, it
* can be useful as an error check at cleanup time.
*/
static int srcu_readers_active(struct srcu_struct *sp)
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
{
int cpu;
unsigned long sum = 0;
for_each_possible_cpu(cpu) {
sum += ACCESS_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->c[0]);
sum += ACCESS_ONCE(per_cpu_ptr(sp->per_cpu_ref, cpu)->c[1]);
}
return sum;
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
}
/**
* cleanup_srcu_struct - deconstruct a sleep-RCU structure
* @sp: structure to clean up.
*
* Must invoke this after you are finished using a given srcu_struct that
* was initialized via init_srcu_struct(), else you leak memory.
*/
void cleanup_srcu_struct(struct srcu_struct *sp)
{
int sum;
sum = srcu_readers_active(sp);
WARN_ON(sum); /* Leakage unless caller handles error. */
if (sum != 0)
return;
free_percpu(sp->per_cpu_ref);
sp->per_cpu_ref = NULL;
}
EXPORT_SYMBOL_GPL(cleanup_srcu_struct);
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
rcu: Introduce lockdep-based checking to RCU read-side primitives Inspection is proving insufficient to catch all RCU misuses, which is understandable given that rcu_dereference() might be protected by any of four different flavors of RCU (RCU, RCU-bh, RCU-sched, and SRCU), and might also/instead be protected by any of a number of locking primitives. It is therefore time to enlist the aid of lockdep. This set of patches is inspired by earlier work by Peter Zijlstra and Thomas Gleixner, and takes the following approach: o Set up separate lockdep classes for RCU, RCU-bh, and RCU-sched. o Set up separate lockdep classes for each instance of SRCU. o Create primitives that check for being in an RCU read-side critical section. These return exact answers if lockdep is fully enabled, but if unsure, report being in an RCU read-side critical section. (We want to avoid false positives!) The primitives are: For RCU: rcu_read_lock_held(void) For RCU-bh: rcu_read_lock_bh_held(void) For RCU-sched: rcu_read_lock_sched_held(void) For SRCU: srcu_read_lock_held(struct srcu_struct *sp) o Add rcu_dereference_check(), which takes a second argument in which one places a boolean expression based on the above primitives and/or lockdep_is_held(). o A new kernel configuration parameter, CONFIG_PROVE_RCU, enables rcu_dereference_check(). This depends on CONFIG_PROVE_LOCKING, and should be quite helpful during the transition period while CONFIG_PROVE_RCU-unaware patches are in flight. The existing rcu_dereference() primitive does no checking, but upcoming patches will change that. Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1266887105-1528-1-git-send-email-paulmck@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-23 02:04:45 +01:00
/*
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
* Counts the new reader in the appropriate per-CPU element of the
* srcu_struct. Must be called from process context.
* Returns an index that must be passed to the matching srcu_read_unlock().
*/
rcu: Introduce lockdep-based checking to RCU read-side primitives Inspection is proving insufficient to catch all RCU misuses, which is understandable given that rcu_dereference() might be protected by any of four different flavors of RCU (RCU, RCU-bh, RCU-sched, and SRCU), and might also/instead be protected by any of a number of locking primitives. It is therefore time to enlist the aid of lockdep. This set of patches is inspired by earlier work by Peter Zijlstra and Thomas Gleixner, and takes the following approach: o Set up separate lockdep classes for RCU, RCU-bh, and RCU-sched. o Set up separate lockdep classes for each instance of SRCU. o Create primitives that check for being in an RCU read-side critical section. These return exact answers if lockdep is fully enabled, but if unsure, report being in an RCU read-side critical section. (We want to avoid false positives!) The primitives are: For RCU: rcu_read_lock_held(void) For RCU-bh: rcu_read_lock_bh_held(void) For RCU-sched: rcu_read_lock_sched_held(void) For SRCU: srcu_read_lock_held(struct srcu_struct *sp) o Add rcu_dereference_check(), which takes a second argument in which one places a boolean expression based on the above primitives and/or lockdep_is_held(). o A new kernel configuration parameter, CONFIG_PROVE_RCU, enables rcu_dereference_check(). This depends on CONFIG_PROVE_LOCKING, and should be quite helpful during the transition period while CONFIG_PROVE_RCU-unaware patches are in flight. The existing rcu_dereference() primitive does no checking, but upcoming patches will change that. Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1266887105-1528-1-git-send-email-paulmck@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-23 02:04:45 +01:00
int __srcu_read_lock(struct srcu_struct *sp)
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
{
int idx;
preempt_disable();
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
idx = rcu_dereference_index_check(sp->completed,
rcu_read_lock_sched_held()) & 0x1;
ACCESS_ONCE(this_cpu_ptr(sp->per_cpu_ref)->c[idx]) += 1;
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
smp_mb(); /* B */ /* Avoid leaking the critical section. */
ACCESS_ONCE(this_cpu_ptr(sp->per_cpu_ref)->seq[idx]) += 1;
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
preempt_enable();
return idx;
}
rcu: Introduce lockdep-based checking to RCU read-side primitives Inspection is proving insufficient to catch all RCU misuses, which is understandable given that rcu_dereference() might be protected by any of four different flavors of RCU (RCU, RCU-bh, RCU-sched, and SRCU), and might also/instead be protected by any of a number of locking primitives. It is therefore time to enlist the aid of lockdep. This set of patches is inspired by earlier work by Peter Zijlstra and Thomas Gleixner, and takes the following approach: o Set up separate lockdep classes for RCU, RCU-bh, and RCU-sched. o Set up separate lockdep classes for each instance of SRCU. o Create primitives that check for being in an RCU read-side critical section. These return exact answers if lockdep is fully enabled, but if unsure, report being in an RCU read-side critical section. (We want to avoid false positives!) The primitives are: For RCU: rcu_read_lock_held(void) For RCU-bh: rcu_read_lock_bh_held(void) For RCU-sched: rcu_read_lock_sched_held(void) For SRCU: srcu_read_lock_held(struct srcu_struct *sp) o Add rcu_dereference_check(), which takes a second argument in which one places a boolean expression based on the above primitives and/or lockdep_is_held(). o A new kernel configuration parameter, CONFIG_PROVE_RCU, enables rcu_dereference_check(). This depends on CONFIG_PROVE_LOCKING, and should be quite helpful during the transition period while CONFIG_PROVE_RCU-unaware patches are in flight. The existing rcu_dereference() primitive does no checking, but upcoming patches will change that. Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1266887105-1528-1-git-send-email-paulmck@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-23 02:04:45 +01:00
EXPORT_SYMBOL_GPL(__srcu_read_lock);
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
rcu: Introduce lockdep-based checking to RCU read-side primitives Inspection is proving insufficient to catch all RCU misuses, which is understandable given that rcu_dereference() might be protected by any of four different flavors of RCU (RCU, RCU-bh, RCU-sched, and SRCU), and might also/instead be protected by any of a number of locking primitives. It is therefore time to enlist the aid of lockdep. This set of patches is inspired by earlier work by Peter Zijlstra and Thomas Gleixner, and takes the following approach: o Set up separate lockdep classes for RCU, RCU-bh, and RCU-sched. o Set up separate lockdep classes for each instance of SRCU. o Create primitives that check for being in an RCU read-side critical section. These return exact answers if lockdep is fully enabled, but if unsure, report being in an RCU read-side critical section. (We want to avoid false positives!) The primitives are: For RCU: rcu_read_lock_held(void) For RCU-bh: rcu_read_lock_bh_held(void) For RCU-sched: rcu_read_lock_sched_held(void) For SRCU: srcu_read_lock_held(struct srcu_struct *sp) o Add rcu_dereference_check(), which takes a second argument in which one places a boolean expression based on the above primitives and/or lockdep_is_held(). o A new kernel configuration parameter, CONFIG_PROVE_RCU, enables rcu_dereference_check(). This depends on CONFIG_PROVE_LOCKING, and should be quite helpful during the transition period while CONFIG_PROVE_RCU-unaware patches are in flight. The existing rcu_dereference() primitive does no checking, but upcoming patches will change that. Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1266887105-1528-1-git-send-email-paulmck@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-23 02:04:45 +01:00
/*
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
* Removes the count for the old reader from the appropriate per-CPU
* element of the srcu_struct. Note that this may well be a different
* CPU than that which was incremented by the corresponding srcu_read_lock().
* Must be called from process context.
*/
rcu: Introduce lockdep-based checking to RCU read-side primitives Inspection is proving insufficient to catch all RCU misuses, which is understandable given that rcu_dereference() might be protected by any of four different flavors of RCU (RCU, RCU-bh, RCU-sched, and SRCU), and might also/instead be protected by any of a number of locking primitives. It is therefore time to enlist the aid of lockdep. This set of patches is inspired by earlier work by Peter Zijlstra and Thomas Gleixner, and takes the following approach: o Set up separate lockdep classes for RCU, RCU-bh, and RCU-sched. o Set up separate lockdep classes for each instance of SRCU. o Create primitives that check for being in an RCU read-side critical section. These return exact answers if lockdep is fully enabled, but if unsure, report being in an RCU read-side critical section. (We want to avoid false positives!) The primitives are: For RCU: rcu_read_lock_held(void) For RCU-bh: rcu_read_lock_bh_held(void) For RCU-sched: rcu_read_lock_sched_held(void) For SRCU: srcu_read_lock_held(struct srcu_struct *sp) o Add rcu_dereference_check(), which takes a second argument in which one places a boolean expression based on the above primitives and/or lockdep_is_held(). o A new kernel configuration parameter, CONFIG_PROVE_RCU, enables rcu_dereference_check(). This depends on CONFIG_PROVE_LOCKING, and should be quite helpful during the transition period while CONFIG_PROVE_RCU-unaware patches are in flight. The existing rcu_dereference() primitive does no checking, but upcoming patches will change that. Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1266887105-1528-1-git-send-email-paulmck@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-23 02:04:45 +01:00
void __srcu_read_unlock(struct srcu_struct *sp, int idx)
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
{
preempt_disable();
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
smp_mb(); /* C */ /* Avoid leaking the critical section. */
2012-02-22 22:29:06 +01:00
ACCESS_ONCE(this_cpu_ptr(sp->per_cpu_ref)->c[idx]) -= 1;
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
preempt_enable();
}
rcu: Introduce lockdep-based checking to RCU read-side primitives Inspection is proving insufficient to catch all RCU misuses, which is understandable given that rcu_dereference() might be protected by any of four different flavors of RCU (RCU, RCU-bh, RCU-sched, and SRCU), and might also/instead be protected by any of a number of locking primitives. It is therefore time to enlist the aid of lockdep. This set of patches is inspired by earlier work by Peter Zijlstra and Thomas Gleixner, and takes the following approach: o Set up separate lockdep classes for RCU, RCU-bh, and RCU-sched. o Set up separate lockdep classes for each instance of SRCU. o Create primitives that check for being in an RCU read-side critical section. These return exact answers if lockdep is fully enabled, but if unsure, report being in an RCU read-side critical section. (We want to avoid false positives!) The primitives are: For RCU: rcu_read_lock_held(void) For RCU-bh: rcu_read_lock_bh_held(void) For RCU-sched: rcu_read_lock_sched_held(void) For SRCU: srcu_read_lock_held(struct srcu_struct *sp) o Add rcu_dereference_check(), which takes a second argument in which one places a boolean expression based on the above primitives and/or lockdep_is_held(). o A new kernel configuration parameter, CONFIG_PROVE_RCU, enables rcu_dereference_check(). This depends on CONFIG_PROVE_LOCKING, and should be quite helpful during the transition period while CONFIG_PROVE_RCU-unaware patches are in flight. The existing rcu_dereference() primitive does no checking, but upcoming patches will change that. Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: laijs@cn.fujitsu.com Cc: dipankar@in.ibm.com Cc: mathieu.desnoyers@polymtl.ca Cc: josh@joshtriplett.org Cc: dvhltc@us.ibm.com Cc: niv@us.ibm.com Cc: peterz@infradead.org Cc: rostedt@goodmis.org Cc: Valdis.Kletnieks@vt.edu Cc: dhowells@redhat.com LKML-Reference: <1266887105-1528-1-git-send-email-paulmck@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2010-02-23 02:04:45 +01:00
EXPORT_SYMBOL_GPL(__srcu_read_unlock);
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
/*
* We use an adaptive strategy for synchronize_srcu() and especially for
* synchronize_srcu_expedited(). We spin for a fixed time period
* (defined below) to allow SRCU readers to exit their read-side critical
* sections. If there are still some readers after 10 microseconds,
* we repeatedly block for 1-millisecond time periods. This approach
* has done well in testing, so there is no need for a config parameter.
*/
#define SRCU_RETRY_CHECK_DELAY 5
#define SYNCHRONIZE_SRCU_TRYCOUNT 2
#define SYNCHRONIZE_SRCU_EXP_TRYCOUNT 12
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
/*
* @@@ Wait until all pre-existing readers complete. Such readers
* will have used the index specified by "idx".
* the caller should ensures the ->completed is not changed while checking
* and idx = (->completed & 1) ^ 1
*/
static bool try_check_zero(struct srcu_struct *sp, int idx, int trycount)
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
{
for (;;) {
if (srcu_readers_active_idx_check(sp, idx))
return true;
if (--trycount <= 0)
return false;
udelay(SRCU_RETRY_CHECK_DELAY);
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
}
}
/*
* Increment the ->completed counter so that future SRCU readers will
* use the other rank of the ->c[] and ->seq[] arrays. This allows
* us to wait for pre-existing readers in a starvation-free manner.
*/
static void srcu_flip(struct srcu_struct *sp)
{
sp->completed++;
}
/*
* Enqueue an SRCU callback on the specified srcu_struct structure,
* initiating grace-period processing if it is not already running.
*/
void call_srcu(struct srcu_struct *sp, struct rcu_head *head,
void (*func)(struct rcu_head *head))
{
unsigned long flags;
head->next = NULL;
head->func = func;
spin_lock_irqsave(&sp->queue_lock, flags);
rcu_batch_queue(&sp->batch_queue, head);
if (!sp->running) {
sp->running = true;
queue_delayed_work(system_nrt_wq, &sp->work, 0);
}
spin_unlock_irqrestore(&sp->queue_lock, flags);
}
EXPORT_SYMBOL_GPL(call_srcu);
struct rcu_synchronize {
struct rcu_head head;
struct completion completion;
};
/*
* Awaken the corresponding synchronize_srcu() instance now that a
* grace period has elapsed.
*/
static void wakeme_after_rcu(struct rcu_head *head)
{
struct rcu_synchronize *rcu;
rcu = container_of(head, struct rcu_synchronize, head);
complete(&rcu->completion);
}
static void srcu_advance_batches(struct srcu_struct *sp, int trycount);
static void srcu_reschedule(struct srcu_struct *sp);
/*
* Helper function for synchronize_srcu() and synchronize_srcu_expedited().
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
*/
static void __synchronize_srcu(struct srcu_struct *sp, int trycount)
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
{
struct rcu_synchronize rcu;
struct rcu_head *head = &rcu.head;
bool done = false;
rcu_lockdep_assert(!lock_is_held(&sp->dep_map) &&
!lock_is_held(&rcu_bh_lock_map) &&
!lock_is_held(&rcu_lock_map) &&
!lock_is_held(&rcu_sched_lock_map),
"Illegal synchronize_srcu() in same-type SRCU (or RCU) read-side critical section");
init_completion(&rcu.completion);
head->next = NULL;
head->func = wakeme_after_rcu;
spin_lock_irq(&sp->queue_lock);
if (!sp->running) {
/* steal the processing owner */
sp->running = true;
rcu_batch_queue(&sp->batch_check0, head);
spin_unlock_irq(&sp->queue_lock);
srcu_advance_batches(sp, trycount);
if (!rcu_batch_empty(&sp->batch_done)) {
BUG_ON(sp->batch_done.head != head);
rcu_batch_dequeue(&sp->batch_done);
done = true;
}
/* give the processing owner to work_struct */
srcu_reschedule(sp);
} else {
rcu_batch_queue(&sp->batch_queue, head);
spin_unlock_irq(&sp->queue_lock);
}
if (!done)
wait_for_completion(&rcu.completion);
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
}
/**
* synchronize_srcu - wait for prior SRCU read-side critical-section completion
* @sp: srcu_struct with which to synchronize.
*
* Flip the completed counter, and wait for the old count to drain to zero.
* As with classic RCU, the updater must use some separate means of
* synchronizing concurrent updates. Can block; must be called from
* process context.
*
* Note that it is illegal to call synchronize_srcu() from the corresponding
* SRCU read-side critical section; doing so will result in deadlock.
* However, it is perfectly legal to call synchronize_srcu() on one
* srcu_struct from some other srcu_struct's read-side critical section.
*/
void synchronize_srcu(struct srcu_struct *sp)
{
__synchronize_srcu(sp, SYNCHRONIZE_SRCU_TRYCOUNT);
}
EXPORT_SYMBOL_GPL(synchronize_srcu);
/**
* synchronize_srcu_expedited - Brute-force SRCU grace period
* @sp: srcu_struct with which to synchronize.
*
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
* Wait for an SRCU grace period to elapse, but be more aggressive about
* spinning rather than blocking when waiting.
*
* Note that it is illegal to call this function while holding any lock
rcu: Direct algorithmic SRCU implementation The current implementation of synchronize_srcu_expedited() can cause severe OS jitter due to its use of synchronize_sched(), which in turn invokes try_stop_cpus(), which causes each CPU to be sent an IPI. This can result in severe performance degradation for real-time workloads and especially for short-interation-length HPC workloads. Furthermore, because only one instance of try_stop_cpus() can be making forward progress at a given time, only one instance of synchronize_srcu_expedited() can make forward progress at a time, even if they are all operating on distinct srcu_struct structures. This commit, inspired by an earlier implementation by Peter Zijlstra (https://lkml.org/lkml/2012/1/31/211) and by further offline discussions, takes a strictly algorithmic bits-in-memory approach. This has the disadvantage of requiring one explicit memory-barrier instruction in each of srcu_read_lock() and srcu_read_unlock(), but on the other hand completely dispenses with OS jitter and furthermore allows SRCU to be used freely by CPUs that RCU believes to be idle or offline. The update-side implementation handles the single read-side memory barrier by rechecking the per-CPU counters after summing them and by running through the update-side state machine twice. This implementation has passed moderate rcutorture testing on both x86 and Power. Also updated to use this_cpu_ptr() instead of per_cpu_ptr(), as suggested by Peter Zijlstra. Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
2012-02-05 16:42:44 +01:00
* that is acquired by a CPU-hotplug notifier. It is also illegal to call
* synchronize_srcu_expedited() from the corresponding SRCU read-side
* critical section; doing so will result in deadlock. However, it is
* perfectly legal to call synchronize_srcu_expedited() on one srcu_struct
* from some other srcu_struct's read-side critical section, as long as
* the resulting graph of srcu_structs is acyclic.
*/
void synchronize_srcu_expedited(struct srcu_struct *sp)
{
__synchronize_srcu(sp, SYNCHRONIZE_SRCU_EXP_TRYCOUNT);
}
EXPORT_SYMBOL_GPL(synchronize_srcu_expedited);
/**
* srcu_barrier - Wait until all in-flight call_srcu() callbacks complete.
*/
void srcu_barrier(struct srcu_struct *sp)
{
synchronize_srcu(sp);
}
EXPORT_SYMBOL_GPL(srcu_barrier);
[PATCH] srcu-3: RCU variant permitting read-side blocking Updated patch adding a variant of RCU that permits sleeping in read-side critical sections. SRCU is as follows: o Each use of SRCU creates its own srcu_struct, and each srcu_struct has its own set of grace periods. This is critical, as it prevents one subsystem with a blocking reader from holding up SRCU grace periods for other subsystems. o The SRCU primitives (srcu_read_lock(), srcu_read_unlock(), and synchronize_srcu()) all take a pointer to a srcu_struct. o The SRCU primitives must be called from process context. o srcu_read_lock() returns an int that must be passed to the matching srcu_read_unlock(). Realtime RCU avoids the need for this by storing the state in the task struct, but SRCU needs to allow a given code path to pass through multiple SRCU domains -- storing state in the task struct would therefore require either arbitrary space in the task struct or arbitrary limits on SRCU nesting. So I kicked the state-storage problem up to the caller. Of course, it is not permitted to call synchronize_srcu() while in an SRCU read-side critical section. o There is no call_srcu(). It would not be hard to implement one, but it seems like too easy a way to OOM the system. (Hey, we have enough trouble with call_rcu(), which does -not- permit readers to sleep!!!) So, if you want it, please tell me why... [josht@us.ibm.com: sparse notation] Signed-off-by: Paul E. McKenney <paulmck@us.ibm.com> Signed-off-by: Josh Triplett <josh@freedesktop.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 11:17:02 +02:00
/**
* srcu_batches_completed - return batches completed.
* @sp: srcu_struct on which to report batch completion.
*
* Report the number of batches, correlated with, but not necessarily
* precisely the same as, the number of grace periods that have elapsed.
*/
long srcu_batches_completed(struct srcu_struct *sp)
{
return sp->completed;
}
EXPORT_SYMBOL_GPL(srcu_batches_completed);
#define SRCU_CALLBACK_BATCH 10
#define SRCU_INTERVAL 1
/*
* Move any new SRCU callbacks to the first stage of the SRCU grace
* period pipeline.
*/
static void srcu_collect_new(struct srcu_struct *sp)
{
if (!rcu_batch_empty(&sp->batch_queue)) {
spin_lock_irq(&sp->queue_lock);
rcu_batch_move(&sp->batch_check0, &sp->batch_queue);
spin_unlock_irq(&sp->queue_lock);
}
}
/*
* Core SRCU state machine. Advance callbacks from ->batch_check0 to
* ->batch_check1 and then to ->batch_done as readers drain.
*/
static void srcu_advance_batches(struct srcu_struct *sp, int trycount)
{
int idx = 1 ^ (sp->completed & 1);
/*
* Because readers might be delayed for an extended period after
* fetching ->completed for their index, at any point in time there
* might well be readers using both idx=0 and idx=1. We therefore
* need to wait for readers to clear from both index values before
* invoking a callback.
*/
if (rcu_batch_empty(&sp->batch_check0) &&
rcu_batch_empty(&sp->batch_check1))
return; /* no callbacks need to be advanced */
if (!try_check_zero(sp, idx, trycount))
return; /* failed to advance, will try after SRCU_INTERVAL */
/*
* The callbacks in ->batch_check1 have already done with their
* first zero check and flip back when they were enqueued on
* ->batch_check0 in a previous invocation of srcu_advance_batches().
* (Presumably try_check_zero() returned false during that
* invocation, leaving the callbacks stranded on ->batch_check1.)
* They are therefore ready to invoke, so move them to ->batch_done.
*/
rcu_batch_move(&sp->batch_done, &sp->batch_check1);
if (rcu_batch_empty(&sp->batch_check0))
return; /* no callbacks need to be advanced */
srcu_flip(sp);
/*
* The callbacks in ->batch_check0 just finished their
* first check zero and flip, so move them to ->batch_check1
* for future checking on the other idx.
*/
rcu_batch_move(&sp->batch_check1, &sp->batch_check0);
/*
* SRCU read-side critical sections are normally short, so check
* at least twice in quick succession after a flip.
*/
trycount = trycount < 2 ? 2 : trycount;
if (!try_check_zero(sp, idx^1, trycount))
return; /* failed to advance, will try after SRCU_INTERVAL */
/*
* The callbacks in ->batch_check1 have now waited for all
* pre-existing readers using both idx values. They are therefore
* ready to invoke, so move them to ->batch_done.
*/
rcu_batch_move(&sp->batch_done, &sp->batch_check1);
}
/*
* Invoke a limited number of SRCU callbacks that have passed through
* their grace period. If there are more to do, SRCU will reschedule
* the workqueue.
*/
static void srcu_invoke_callbacks(struct srcu_struct *sp)
{
int i;
struct rcu_head *head;
for (i = 0; i < SRCU_CALLBACK_BATCH; i++) {
head = rcu_batch_dequeue(&sp->batch_done);
if (!head)
break;
local_bh_disable();
head->func(head);
local_bh_enable();
}
}
/*
* Finished one round of SRCU grace period. Start another if there are
* more SRCU callbacks queued, otherwise put SRCU into not-running state.
*/
static void srcu_reschedule(struct srcu_struct *sp)
{
bool pending = true;
if (rcu_batch_empty(&sp->batch_done) &&
rcu_batch_empty(&sp->batch_check1) &&
rcu_batch_empty(&sp->batch_check0) &&
rcu_batch_empty(&sp->batch_queue)) {
spin_lock_irq(&sp->queue_lock);
if (rcu_batch_empty(&sp->batch_done) &&
rcu_batch_empty(&sp->batch_check1) &&
rcu_batch_empty(&sp->batch_check0) &&
rcu_batch_empty(&sp->batch_queue)) {
sp->running = false;
pending = false;
}
spin_unlock_irq(&sp->queue_lock);
}
if (pending)
queue_delayed_work(system_nrt_wq, &sp->work, SRCU_INTERVAL);
}
/*
* This is the work-queue function that handles SRCU grace periods.
*/
static void process_srcu(struct work_struct *work)
{
struct srcu_struct *sp;
sp = container_of(work, struct srcu_struct, work.work);
srcu_collect_new(sp);
srcu_advance_batches(sp, 1);
srcu_invoke_callbacks(sp);
srcu_reschedule(sp);
}