rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
/*
|
|
|
|
* Read-Copy Update mechanism for mutual exclusion (tree-based version)
|
|
|
|
* Internal non-public definitions that provide either classic
|
2011-03-02 22:15:15 +01:00
|
|
|
* or preemptible semantics.
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
*
|
|
|
|
* 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 Red Hat, 2009
|
|
|
|
* Copyright IBM Corporation, 2009
|
|
|
|
*
|
|
|
|
* Author: Ingo Molnar <mingo@elte.hu>
|
|
|
|
* Paul E. McKenney <paulmck@linux.vnet.ibm.com>
|
|
|
|
*/
|
|
|
|
|
2009-12-02 21:10:15 +01:00
|
|
|
#include <linux/delay.h>
|
2010-10-21 05:29:05 +02:00
|
|
|
#include <linux/stop_machine.h>
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
|
2010-04-13 23:19:23 +02:00
|
|
|
/*
|
|
|
|
* Check the RCU kernel configuration parameters and print informative
|
|
|
|
* messages about anything out of the ordinary. If you like #ifdef, you
|
|
|
|
* will love this function.
|
|
|
|
*/
|
|
|
|
static void __init rcu_bootup_announce_oddness(void)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_RCU_TRACE
|
|
|
|
printk(KERN_INFO "\tRCU debugfs-based tracing is enabled.\n");
|
|
|
|
#endif
|
|
|
|
#if (defined(CONFIG_64BIT) && CONFIG_RCU_FANOUT != 64) || (!defined(CONFIG_64BIT) && CONFIG_RCU_FANOUT != 32)
|
|
|
|
printk(KERN_INFO "\tCONFIG_RCU_FANOUT set to non-default value of %d\n",
|
|
|
|
CONFIG_RCU_FANOUT);
|
|
|
|
#endif
|
|
|
|
#ifdef CONFIG_RCU_FANOUT_EXACT
|
|
|
|
printk(KERN_INFO "\tHierarchical RCU autobalancing is disabled.\n");
|
|
|
|
#endif
|
|
|
|
#ifdef CONFIG_RCU_FAST_NO_HZ
|
|
|
|
printk(KERN_INFO
|
|
|
|
"\tRCU dyntick-idle grace-period acceleration is enabled.\n");
|
|
|
|
#endif
|
|
|
|
#ifdef CONFIG_PROVE_RCU
|
|
|
|
printk(KERN_INFO "\tRCU lockdep checking is enabled.\n");
|
|
|
|
#endif
|
|
|
|
#ifdef CONFIG_RCU_TORTURE_TEST_RUNNABLE
|
|
|
|
printk(KERN_INFO "\tRCU torture testing starts during boot.\n");
|
|
|
|
#endif
|
2010-08-30 18:52:50 +02:00
|
|
|
#if defined(CONFIG_TREE_PREEMPT_RCU) && !defined(CONFIG_RCU_CPU_STALL_VERBOSE)
|
2010-04-13 23:19:23 +02:00
|
|
|
printk(KERN_INFO "\tVerbose stalled-CPUs detection is disabled.\n");
|
|
|
|
#endif
|
|
|
|
#if NUM_RCU_LVL_4 != 0
|
|
|
|
printk(KERN_INFO "\tExperimental four-level hierarchy is enabled.\n");
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
#ifdef CONFIG_TREE_PREEMPT_RCU
|
|
|
|
|
|
|
|
struct rcu_state rcu_preempt_state = RCU_STATE_INITIALIZER(rcu_preempt_state);
|
|
|
|
DEFINE_PER_CPU(struct rcu_data, rcu_preempt_data);
|
2011-02-07 21:47:15 +01:00
|
|
|
static struct rcu_state *rcu_state = &rcu_preempt_state;
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
|
2009-12-02 21:10:15 +01:00
|
|
|
static int rcu_preempted_readers_exp(struct rcu_node *rnp);
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
/*
|
|
|
|
* Tell them what RCU they are running.
|
|
|
|
*/
|
2009-11-11 20:28:06 +01:00
|
|
|
static void __init rcu_bootup_announce(void)
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
{
|
2011-03-02 22:15:15 +01:00
|
|
|
printk(KERN_INFO "Preemptible hierarchical RCU implementation.\n");
|
2010-04-13 23:19:23 +02:00
|
|
|
rcu_bootup_announce_oddness();
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Return the number of RCU-preempt batches processed thus far
|
|
|
|
* for debug and statistics.
|
|
|
|
*/
|
|
|
|
long rcu_batches_completed_preempt(void)
|
|
|
|
{
|
|
|
|
return rcu_preempt_state.completed;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(rcu_batches_completed_preempt);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Return the number of RCU batches processed thus far for debug & stats.
|
|
|
|
*/
|
|
|
|
long rcu_batches_completed(void)
|
|
|
|
{
|
|
|
|
return rcu_batches_completed_preempt();
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(rcu_batches_completed);
|
|
|
|
|
2010-01-05 00:09:10 +01:00
|
|
|
/*
|
|
|
|
* Force a quiescent state for preemptible RCU.
|
|
|
|
*/
|
|
|
|
void rcu_force_quiescent_state(void)
|
|
|
|
{
|
|
|
|
force_quiescent_state(&rcu_preempt_state, 0);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(rcu_force_quiescent_state);
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Record a preemptible-RCU quiescent state for the specified CPU. Note
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
* that this just means that the task currently running on the CPU is
|
|
|
|
* not in a quiescent state. There might be any number of tasks blocked
|
|
|
|
* while in an RCU read-side critical section.
|
rcu: refactor RCU's context-switch handling
The addition of preemptible RCU to treercu resulted in a bit of
confusion and inefficiency surrounding the handling of context switches
for RCU-sched and for RCU-preempt. For RCU-sched, a context switch
is a quiescent state, pure and simple, just like it always has been.
For RCU-preempt, a context switch is in no way a quiescent state, but
special handling is required when a task blocks in an RCU read-side
critical section.
However, the callout from the scheduler and the outer loop in ksoftirqd
still calls something named rcu_sched_qs(), whose name is no longer
accurate. Furthermore, when rcu_check_callbacks() notes an RCU-sched
quiescent state, it ends up unnecessarily (though harmlessly, aside
from the performance hit) enqueuing the current task if it happens to
be running in an RCU-preempt read-side critical section. This not only
increases the maximum latency of scheduler_tick(), it also needlessly
increases the overhead of the next outermost rcu_read_unlock() invocation.
This patch addresses this situation by separating the notion of RCU's
context-switch handling from that of RCU-sched's quiescent states.
The context-switch handling is covered by rcu_note_context_switch() in
general and by rcu_preempt_note_context_switch() for preemptible RCU.
This permits rcu_sched_qs() to handle quiescent states and only quiescent
states. It also reduces the maximum latency of scheduler_tick(), though
probably by much less than a microsecond. Finally, it means that tasks
within preemptible-RCU read-side critical sections avoid incurring the
overhead of queuing unless there really is a context switch.
Suggested-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Acked-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
2010-04-02 02:37:01 +02:00
|
|
|
*
|
|
|
|
* Unlike the other rcu_*_qs() functions, callers to this function
|
|
|
|
* must disable irqs in order to protect the assignment to
|
|
|
|
* ->rcu_read_unlock_special.
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
*/
|
2009-09-13 18:15:10 +02:00
|
|
|
static void rcu_preempt_qs(int cpu)
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
{
|
|
|
|
struct rcu_data *rdp = &per_cpu(rcu_preempt_data, cpu);
|
rcu: refactor RCU's context-switch handling
The addition of preemptible RCU to treercu resulted in a bit of
confusion and inefficiency surrounding the handling of context switches
for RCU-sched and for RCU-preempt. For RCU-sched, a context switch
is a quiescent state, pure and simple, just like it always has been.
For RCU-preempt, a context switch is in no way a quiescent state, but
special handling is required when a task blocks in an RCU read-side
critical section.
However, the callout from the scheduler and the outer loop in ksoftirqd
still calls something named rcu_sched_qs(), whose name is no longer
accurate. Furthermore, when rcu_check_callbacks() notes an RCU-sched
quiescent state, it ends up unnecessarily (though harmlessly, aside
from the performance hit) enqueuing the current task if it happens to
be running in an RCU-preempt read-side critical section. This not only
increases the maximum latency of scheduler_tick(), it also needlessly
increases the overhead of the next outermost rcu_read_unlock() invocation.
This patch addresses this situation by separating the notion of RCU's
context-switch handling from that of RCU-sched's quiescent states.
The context-switch handling is covered by rcu_note_context_switch() in
general and by rcu_preempt_note_context_switch() for preemptible RCU.
This permits rcu_sched_qs() to handle quiescent states and only quiescent
states. It also reduces the maximum latency of scheduler_tick(), though
probably by much less than a microsecond. Finally, it means that tasks
within preemptible-RCU read-side critical sections avoid incurring the
overhead of queuing unless there really is a context switch.
Suggested-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Acked-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
2010-04-02 02:37:01 +02:00
|
|
|
|
rcu: Simplify association of quiescent states with grace periods
The rdp->passed_quiesc_completed fields are used to properly
associate the recorded quiescent state with a grace period. It
is OK to wrongly associate a given quiescent state with a
preceding grace period, but it is fatal to associate a given
quiescent state with a grace period that begins after the
quiescent state occurred. Grace periods are numbered, and the
following fields track them:
o ->gpnum is the number of the grace period currently in
progress, or the number of the last grace period to
complete if no grace period is currently in progress.
o ->completed is the number of the last grace period to
have completed.
These two fields are equal if there is no grace period in
progress, otherwise ->gpnum is one greater than ->completed.
But the rdp->passed_quiesc_completed field compared against
->completed, and if equal, the quiescent state is presumed to
count against the current grace period.
The earlier code copied rdp->completed to
rdp->passed_quiesc_completed, which has been made to work, but
is error-prone. In contrast, copying one less than rdp->gpnum
is guaranteed safe, because rdp->gpnum is not incremented until
after the start of the corresponding grace period. At the end of
the grace period, when ->completed has incremented, then any
quiescent periods recorded previously will be discarded.
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: <12578890421011-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-10 22:37:22 +01:00
|
|
|
rdp->passed_quiesc_completed = rdp->gpnum - 1;
|
2009-09-13 18:15:10 +02:00
|
|
|
barrier();
|
|
|
|
rdp->passed_quiesc = 1;
|
rcu: refactor RCU's context-switch handling
The addition of preemptible RCU to treercu resulted in a bit of
confusion and inefficiency surrounding the handling of context switches
for RCU-sched and for RCU-preempt. For RCU-sched, a context switch
is a quiescent state, pure and simple, just like it always has been.
For RCU-preempt, a context switch is in no way a quiescent state, but
special handling is required when a task blocks in an RCU read-side
critical section.
However, the callout from the scheduler and the outer loop in ksoftirqd
still calls something named rcu_sched_qs(), whose name is no longer
accurate. Furthermore, when rcu_check_callbacks() notes an RCU-sched
quiescent state, it ends up unnecessarily (though harmlessly, aside
from the performance hit) enqueuing the current task if it happens to
be running in an RCU-preempt read-side critical section. This not only
increases the maximum latency of scheduler_tick(), it also needlessly
increases the overhead of the next outermost rcu_read_unlock() invocation.
This patch addresses this situation by separating the notion of RCU's
context-switch handling from that of RCU-sched's quiescent states.
The context-switch handling is covered by rcu_note_context_switch() in
general and by rcu_preempt_note_context_switch() for preemptible RCU.
This permits rcu_sched_qs() to handle quiescent states and only quiescent
states. It also reduces the maximum latency of scheduler_tick(), though
probably by much less than a microsecond. Finally, it means that tasks
within preemptible-RCU read-side critical sections avoid incurring the
overhead of queuing unless there really is a context switch.
Suggested-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Acked-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
2010-04-02 02:37:01 +02:00
|
|
|
current->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS;
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2009-09-13 18:15:10 +02:00
|
|
|
* We have entered the scheduler, and the current task might soon be
|
|
|
|
* context-switched away from. If this task is in an RCU read-side
|
|
|
|
* critical section, we will no longer be able to rely on the CPU to
|
2010-11-30 06:56:39 +01:00
|
|
|
* record that fact, so we enqueue the task on the blkd_tasks list.
|
|
|
|
* The task will dequeue itself when it exits the outermost enclosing
|
|
|
|
* RCU read-side critical section. Therefore, the current grace period
|
|
|
|
* cannot be permitted to complete until the blkd_tasks list entries
|
|
|
|
* predating the current grace period drain, in other words, until
|
|
|
|
* rnp->gp_tasks becomes NULL.
|
2009-09-13 18:15:10 +02:00
|
|
|
*
|
|
|
|
* Caller must disable preemption.
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
*/
|
2009-09-13 18:15:10 +02:00
|
|
|
static void rcu_preempt_note_context_switch(int cpu)
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
{
|
|
|
|
struct task_struct *t = current;
|
2009-09-13 18:15:10 +02:00
|
|
|
unsigned long flags;
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
struct rcu_data *rdp;
|
|
|
|
struct rcu_node *rnp;
|
|
|
|
|
|
|
|
if (t->rcu_read_lock_nesting &&
|
|
|
|
(t->rcu_read_unlock_special & RCU_READ_UNLOCK_BLOCKED) == 0) {
|
|
|
|
|
|
|
|
/* Possibly blocking in an RCU read-side critical section. */
|
2010-06-28 10:25:04 +02:00
|
|
|
rdp = per_cpu_ptr(rcu_preempt_state.rda, cpu);
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
rnp = rdp->mynode;
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_lock_irqsave(&rnp->lock, flags);
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BLOCKED;
|
2009-08-28 00:00:12 +02:00
|
|
|
t->rcu_blocked_node = rnp;
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* If this CPU has already checked in, then this task
|
|
|
|
* will hold up the next grace period rather than the
|
|
|
|
* current grace period. Queue the task accordingly.
|
|
|
|
* If the task is queued for the current grace period
|
|
|
|
* (i.e., this CPU has not yet passed through a quiescent
|
|
|
|
* state for the current grace period), then as long
|
|
|
|
* as that task remains queued, the current grace period
|
2010-11-30 06:56:39 +01:00
|
|
|
* cannot end. Note that there is some uncertainty as
|
|
|
|
* to exactly when the current grace period started.
|
|
|
|
* We take a conservative approach, which can result
|
|
|
|
* in unnecessarily waiting on tasks that started very
|
|
|
|
* slightly after the current grace period began. C'est
|
|
|
|
* la vie!!!
|
2009-09-13 18:15:09 +02:00
|
|
|
*
|
|
|
|
* But first, note that the current CPU must still be
|
|
|
|
* on line!
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
*/
|
2009-09-13 18:15:09 +02:00
|
|
|
WARN_ON_ONCE((rdp->grpmask & rnp->qsmaskinit) == 0);
|
2009-09-18 18:50:18 +02:00
|
|
|
WARN_ON_ONCE(!list_empty(&t->rcu_node_entry));
|
2010-11-30 06:56:39 +01:00
|
|
|
if ((rnp->qsmask & rdp->grpmask) && rnp->gp_tasks != NULL) {
|
|
|
|
list_add(&t->rcu_node_entry, rnp->gp_tasks->prev);
|
|
|
|
rnp->gp_tasks = &t->rcu_node_entry;
|
2011-02-07 21:47:15 +01:00
|
|
|
#ifdef CONFIG_RCU_BOOST
|
|
|
|
if (rnp->boost_tasks != NULL)
|
|
|
|
rnp->boost_tasks = rnp->gp_tasks;
|
|
|
|
#endif /* #ifdef CONFIG_RCU_BOOST */
|
2010-11-30 06:56:39 +01:00
|
|
|
} else {
|
|
|
|
list_add(&t->rcu_node_entry, &rnp->blkd_tasks);
|
|
|
|
if (rnp->qsmask & rdp->grpmask)
|
|
|
|
rnp->gp_tasks = &t->rcu_node_entry;
|
|
|
|
}
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_unlock_irqrestore(&rnp->lock, flags);
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Either we were not in an RCU read-side critical section to
|
|
|
|
* begin with, or we have now recorded that critical section
|
|
|
|
* globally. Either way, we can now note a quiescent state
|
|
|
|
* for this CPU. Again, if we were in an RCU read-side critical
|
|
|
|
* section, and if that critical section was blocking the current
|
|
|
|
* grace period, then the fact that the task has been enqueued
|
|
|
|
* means that we continue to block the current grace period.
|
|
|
|
*/
|
2009-09-18 18:50:18 +02:00
|
|
|
local_irq_save(flags);
|
rcu: refactor RCU's context-switch handling
The addition of preemptible RCU to treercu resulted in a bit of
confusion and inefficiency surrounding the handling of context switches
for RCU-sched and for RCU-preempt. For RCU-sched, a context switch
is a quiescent state, pure and simple, just like it always has been.
For RCU-preempt, a context switch is in no way a quiescent state, but
special handling is required when a task blocks in an RCU read-side
critical section.
However, the callout from the scheduler and the outer loop in ksoftirqd
still calls something named rcu_sched_qs(), whose name is no longer
accurate. Furthermore, when rcu_check_callbacks() notes an RCU-sched
quiescent state, it ends up unnecessarily (though harmlessly, aside
from the performance hit) enqueuing the current task if it happens to
be running in an RCU-preempt read-side critical section. This not only
increases the maximum latency of scheduler_tick(), it also needlessly
increases the overhead of the next outermost rcu_read_unlock() invocation.
This patch addresses this situation by separating the notion of RCU's
context-switch handling from that of RCU-sched's quiescent states.
The context-switch handling is covered by rcu_note_context_switch() in
general and by rcu_preempt_note_context_switch() for preemptible RCU.
This permits rcu_sched_qs() to handle quiescent states and only quiescent
states. It also reduces the maximum latency of scheduler_tick(), though
probably by much less than a microsecond. Finally, it means that tasks
within preemptible-RCU read-side critical sections avoid incurring the
overhead of queuing unless there really is a context switch.
Suggested-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Acked-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
2010-04-02 02:37:01 +02:00
|
|
|
rcu_preempt_qs(cpu);
|
2009-09-18 18:50:18 +02:00
|
|
|
local_irq_restore(flags);
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Tree-preemptible RCU implementation for rcu_read_lock().
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
* Just increment ->rcu_read_lock_nesting, shared state will be updated
|
|
|
|
* if we block.
|
|
|
|
*/
|
|
|
|
void __rcu_read_lock(void)
|
|
|
|
{
|
2010-08-20 01:57:45 +02:00
|
|
|
current->rcu_read_lock_nesting++;
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
barrier(); /* needed if we ever invoke rcu_read_lock in rcutree.c */
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(__rcu_read_lock);
|
|
|
|
|
2009-09-23 18:50:41 +02:00
|
|
|
/*
|
|
|
|
* Check for preempted RCU readers blocking the current grace period
|
|
|
|
* for the specified rcu_node structure. If the caller needs a reliable
|
|
|
|
* answer, it must hold the rcu_node's ->lock.
|
|
|
|
*/
|
2011-02-07 21:47:15 +01:00
|
|
|
static int rcu_preempt_blocked_readers_cgp(struct rcu_node *rnp)
|
2009-09-23 18:50:41 +02:00
|
|
|
{
|
2010-11-30 06:56:39 +01:00
|
|
|
return rnp->gp_tasks != NULL;
|
2009-09-23 18:50:41 +02:00
|
|
|
}
|
|
|
|
|
rcu: Fix grace-period-stall bug on large systems with CPU hotplug
When the last CPU of a given leaf rcu_node structure goes
offline, all of the tasks queued on that leaf rcu_node structure
(due to having blocked in their current RCU read-side critical
sections) are requeued onto the root rcu_node structure. This
requeuing is carried out by rcu_preempt_offline_tasks().
However, it is possible that these queued tasks are the only
thing preventing the leaf rcu_node structure from reporting a
quiescent state up the rcu_node hierarchy. Unfortunately, the
old code would fail to do this reporting, resulting in a
grace-period stall given the following sequence of events:
1. Kernel built for more than 32 CPUs on 32-bit systems or for more
than 64 CPUs on 64-bit systems, so that there is more than one
rcu_node structure. (Or CONFIG_RCU_FANOUT is artificially set
to a number smaller than CONFIG_NR_CPUS.)
2. The kernel is built with CONFIG_TREE_PREEMPT_RCU.
3. A task running on a CPU associated with a given leaf rcu_node
structure blocks while in an RCU read-side critical section
-and- that CPU has not yet passed through a quiescent state
for the current RCU grace period. This will cause the task
to be queued on the leaf rcu_node's blocked_tasks[] array, in
particular, on the element of this array corresponding to the
current grace period.
4. Each of the remaining CPUs corresponding to this same leaf rcu_node
structure pass through a quiescent state. However, the task is
still in its RCU read-side critical section, so these quiescent
states cannot be reported further up the rcu_node hierarchy.
Nevertheless, all bits in the leaf rcu_node structure's ->qsmask
field are now zero.
5. Each of the remaining CPUs go offline. (The events in step
#4 and #5 can happen in any order as long as each CPU passes
through a quiescent state before going offline.)
6. When the last CPU goes offline, __rcu_offline_cpu() will invoke
rcu_preempt_offline_tasks(), which will move the task to the
root rcu_node structure, but without reporting a quiescent state
up the rcu_node hierarchy (and this failure to report a quiescent
state is the bug).
But because this leaf rcu_node structure's ->qsmask field is
already zero and its ->block_tasks[] entries are all empty,
force_quiescent_state() will skip this rcu_node structure.
Therefore, grace periods are now hung.
This patch abstracts some code out of rcu_read_unlock_special(),
calling the result task_quiet() by analogy with cpu_quiet(), and
invokes task_quiet() from both rcu_read_lock_special() and
__rcu_offline_cpu(). Invoking task_quiet() from
__rcu_offline_cpu() reports the quiescent state up the rcu_node
hierarchy, fixing the bug. This ends up requiring a separate
lock_class_key per level of the rcu_node hierarchy, which this
patch also provides.
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: <12589088301770-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-22 17:53:48 +01:00
|
|
|
/*
|
|
|
|
* Record a quiescent state for all tasks that were previously queued
|
|
|
|
* on the specified rcu_node structure and that were blocking the current
|
|
|
|
* RCU grace period. The caller must hold the specified rnp->lock with
|
|
|
|
* irqs disabled, and this lock is released upon return, but irqs remain
|
|
|
|
* disabled.
|
|
|
|
*/
|
2009-12-02 21:10:13 +01:00
|
|
|
static void rcu_report_unblock_qs_rnp(struct rcu_node *rnp, unsigned long flags)
|
rcu: Fix grace-period-stall bug on large systems with CPU hotplug
When the last CPU of a given leaf rcu_node structure goes
offline, all of the tasks queued on that leaf rcu_node structure
(due to having blocked in their current RCU read-side critical
sections) are requeued onto the root rcu_node structure. This
requeuing is carried out by rcu_preempt_offline_tasks().
However, it is possible that these queued tasks are the only
thing preventing the leaf rcu_node structure from reporting a
quiescent state up the rcu_node hierarchy. Unfortunately, the
old code would fail to do this reporting, resulting in a
grace-period stall given the following sequence of events:
1. Kernel built for more than 32 CPUs on 32-bit systems or for more
than 64 CPUs on 64-bit systems, so that there is more than one
rcu_node structure. (Or CONFIG_RCU_FANOUT is artificially set
to a number smaller than CONFIG_NR_CPUS.)
2. The kernel is built with CONFIG_TREE_PREEMPT_RCU.
3. A task running on a CPU associated with a given leaf rcu_node
structure blocks while in an RCU read-side critical section
-and- that CPU has not yet passed through a quiescent state
for the current RCU grace period. This will cause the task
to be queued on the leaf rcu_node's blocked_tasks[] array, in
particular, on the element of this array corresponding to the
current grace period.
4. Each of the remaining CPUs corresponding to this same leaf rcu_node
structure pass through a quiescent state. However, the task is
still in its RCU read-side critical section, so these quiescent
states cannot be reported further up the rcu_node hierarchy.
Nevertheless, all bits in the leaf rcu_node structure's ->qsmask
field are now zero.
5. Each of the remaining CPUs go offline. (The events in step
#4 and #5 can happen in any order as long as each CPU passes
through a quiescent state before going offline.)
6. When the last CPU goes offline, __rcu_offline_cpu() will invoke
rcu_preempt_offline_tasks(), which will move the task to the
root rcu_node structure, but without reporting a quiescent state
up the rcu_node hierarchy (and this failure to report a quiescent
state is the bug).
But because this leaf rcu_node structure's ->qsmask field is
already zero and its ->block_tasks[] entries are all empty,
force_quiescent_state() will skip this rcu_node structure.
Therefore, grace periods are now hung.
This patch abstracts some code out of rcu_read_unlock_special(),
calling the result task_quiet() by analogy with cpu_quiet(), and
invokes task_quiet() from both rcu_read_lock_special() and
__rcu_offline_cpu(). Invoking task_quiet() from
__rcu_offline_cpu() reports the quiescent state up the rcu_node
hierarchy, fixing the bug. This ends up requiring a separate
lock_class_key per level of the rcu_node hierarchy, which this
patch also provides.
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: <12589088301770-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-22 17:53:48 +01:00
|
|
|
__releases(rnp->lock)
|
|
|
|
{
|
|
|
|
unsigned long mask;
|
|
|
|
struct rcu_node *rnp_p;
|
|
|
|
|
2011-02-07 21:47:15 +01:00
|
|
|
if (rnp->qsmask != 0 || rcu_preempt_blocked_readers_cgp(rnp)) {
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_unlock_irqrestore(&rnp->lock, flags);
|
rcu: Fix grace-period-stall bug on large systems with CPU hotplug
When the last CPU of a given leaf rcu_node structure goes
offline, all of the tasks queued on that leaf rcu_node structure
(due to having blocked in their current RCU read-side critical
sections) are requeued onto the root rcu_node structure. This
requeuing is carried out by rcu_preempt_offline_tasks().
However, it is possible that these queued tasks are the only
thing preventing the leaf rcu_node structure from reporting a
quiescent state up the rcu_node hierarchy. Unfortunately, the
old code would fail to do this reporting, resulting in a
grace-period stall given the following sequence of events:
1. Kernel built for more than 32 CPUs on 32-bit systems or for more
than 64 CPUs on 64-bit systems, so that there is more than one
rcu_node structure. (Or CONFIG_RCU_FANOUT is artificially set
to a number smaller than CONFIG_NR_CPUS.)
2. The kernel is built with CONFIG_TREE_PREEMPT_RCU.
3. A task running on a CPU associated with a given leaf rcu_node
structure blocks while in an RCU read-side critical section
-and- that CPU has not yet passed through a quiescent state
for the current RCU grace period. This will cause the task
to be queued on the leaf rcu_node's blocked_tasks[] array, in
particular, on the element of this array corresponding to the
current grace period.
4. Each of the remaining CPUs corresponding to this same leaf rcu_node
structure pass through a quiescent state. However, the task is
still in its RCU read-side critical section, so these quiescent
states cannot be reported further up the rcu_node hierarchy.
Nevertheless, all bits in the leaf rcu_node structure's ->qsmask
field are now zero.
5. Each of the remaining CPUs go offline. (The events in step
#4 and #5 can happen in any order as long as each CPU passes
through a quiescent state before going offline.)
6. When the last CPU goes offline, __rcu_offline_cpu() will invoke
rcu_preempt_offline_tasks(), which will move the task to the
root rcu_node structure, but without reporting a quiescent state
up the rcu_node hierarchy (and this failure to report a quiescent
state is the bug).
But because this leaf rcu_node structure's ->qsmask field is
already zero and its ->block_tasks[] entries are all empty,
force_quiescent_state() will skip this rcu_node structure.
Therefore, grace periods are now hung.
This patch abstracts some code out of rcu_read_unlock_special(),
calling the result task_quiet() by analogy with cpu_quiet(), and
invokes task_quiet() from both rcu_read_lock_special() and
__rcu_offline_cpu(). Invoking task_quiet() from
__rcu_offline_cpu() reports the quiescent state up the rcu_node
hierarchy, fixing the bug. This ends up requiring a separate
lock_class_key per level of the rcu_node hierarchy, which this
patch also provides.
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: <12589088301770-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-22 17:53:48 +01:00
|
|
|
return; /* Still need more quiescent states! */
|
|
|
|
}
|
|
|
|
|
|
|
|
rnp_p = rnp->parent;
|
|
|
|
if (rnp_p == NULL) {
|
|
|
|
/*
|
|
|
|
* Either there is only one rcu_node in the tree,
|
|
|
|
* or tasks were kicked up to root rcu_node due to
|
|
|
|
* CPUs going offline.
|
|
|
|
*/
|
2009-12-02 21:10:13 +01:00
|
|
|
rcu_report_qs_rsp(&rcu_preempt_state, flags);
|
rcu: Fix grace-period-stall bug on large systems with CPU hotplug
When the last CPU of a given leaf rcu_node structure goes
offline, all of the tasks queued on that leaf rcu_node structure
(due to having blocked in their current RCU read-side critical
sections) are requeued onto the root rcu_node structure. This
requeuing is carried out by rcu_preempt_offline_tasks().
However, it is possible that these queued tasks are the only
thing preventing the leaf rcu_node structure from reporting a
quiescent state up the rcu_node hierarchy. Unfortunately, the
old code would fail to do this reporting, resulting in a
grace-period stall given the following sequence of events:
1. Kernel built for more than 32 CPUs on 32-bit systems or for more
than 64 CPUs on 64-bit systems, so that there is more than one
rcu_node structure. (Or CONFIG_RCU_FANOUT is artificially set
to a number smaller than CONFIG_NR_CPUS.)
2. The kernel is built with CONFIG_TREE_PREEMPT_RCU.
3. A task running on a CPU associated with a given leaf rcu_node
structure blocks while in an RCU read-side critical section
-and- that CPU has not yet passed through a quiescent state
for the current RCU grace period. This will cause the task
to be queued on the leaf rcu_node's blocked_tasks[] array, in
particular, on the element of this array corresponding to the
current grace period.
4. Each of the remaining CPUs corresponding to this same leaf rcu_node
structure pass through a quiescent state. However, the task is
still in its RCU read-side critical section, so these quiescent
states cannot be reported further up the rcu_node hierarchy.
Nevertheless, all bits in the leaf rcu_node structure's ->qsmask
field are now zero.
5. Each of the remaining CPUs go offline. (The events in step
#4 and #5 can happen in any order as long as each CPU passes
through a quiescent state before going offline.)
6. When the last CPU goes offline, __rcu_offline_cpu() will invoke
rcu_preempt_offline_tasks(), which will move the task to the
root rcu_node structure, but without reporting a quiescent state
up the rcu_node hierarchy (and this failure to report a quiescent
state is the bug).
But because this leaf rcu_node structure's ->qsmask field is
already zero and its ->block_tasks[] entries are all empty,
force_quiescent_state() will skip this rcu_node structure.
Therefore, grace periods are now hung.
This patch abstracts some code out of rcu_read_unlock_special(),
calling the result task_quiet() by analogy with cpu_quiet(), and
invokes task_quiet() from both rcu_read_lock_special() and
__rcu_offline_cpu(). Invoking task_quiet() from
__rcu_offline_cpu() reports the quiescent state up the rcu_node
hierarchy, fixing the bug. This ends up requiring a separate
lock_class_key per level of the rcu_node hierarchy, which this
patch also provides.
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: <12589088301770-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-22 17:53:48 +01:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Report up the rest of the hierarchy. */
|
|
|
|
mask = rnp->grpmask;
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */
|
|
|
|
raw_spin_lock(&rnp_p->lock); /* irqs already disabled. */
|
2009-12-02 21:10:13 +01:00
|
|
|
rcu_report_qs_rnp(mask, &rcu_preempt_state, rnp_p, flags);
|
rcu: Fix grace-period-stall bug on large systems with CPU hotplug
When the last CPU of a given leaf rcu_node structure goes
offline, all of the tasks queued on that leaf rcu_node structure
(due to having blocked in their current RCU read-side critical
sections) are requeued onto the root rcu_node structure. This
requeuing is carried out by rcu_preempt_offline_tasks().
However, it is possible that these queued tasks are the only
thing preventing the leaf rcu_node structure from reporting a
quiescent state up the rcu_node hierarchy. Unfortunately, the
old code would fail to do this reporting, resulting in a
grace-period stall given the following sequence of events:
1. Kernel built for more than 32 CPUs on 32-bit systems or for more
than 64 CPUs on 64-bit systems, so that there is more than one
rcu_node structure. (Or CONFIG_RCU_FANOUT is artificially set
to a number smaller than CONFIG_NR_CPUS.)
2. The kernel is built with CONFIG_TREE_PREEMPT_RCU.
3. A task running on a CPU associated with a given leaf rcu_node
structure blocks while in an RCU read-side critical section
-and- that CPU has not yet passed through a quiescent state
for the current RCU grace period. This will cause the task
to be queued on the leaf rcu_node's blocked_tasks[] array, in
particular, on the element of this array corresponding to the
current grace period.
4. Each of the remaining CPUs corresponding to this same leaf rcu_node
structure pass through a quiescent state. However, the task is
still in its RCU read-side critical section, so these quiescent
states cannot be reported further up the rcu_node hierarchy.
Nevertheless, all bits in the leaf rcu_node structure's ->qsmask
field are now zero.
5. Each of the remaining CPUs go offline. (The events in step
#4 and #5 can happen in any order as long as each CPU passes
through a quiescent state before going offline.)
6. When the last CPU goes offline, __rcu_offline_cpu() will invoke
rcu_preempt_offline_tasks(), which will move the task to the
root rcu_node structure, but without reporting a quiescent state
up the rcu_node hierarchy (and this failure to report a quiescent
state is the bug).
But because this leaf rcu_node structure's ->qsmask field is
already zero and its ->block_tasks[] entries are all empty,
force_quiescent_state() will skip this rcu_node structure.
Therefore, grace periods are now hung.
This patch abstracts some code out of rcu_read_unlock_special(),
calling the result task_quiet() by analogy with cpu_quiet(), and
invokes task_quiet() from both rcu_read_lock_special() and
__rcu_offline_cpu(). Invoking task_quiet() from
__rcu_offline_cpu() reports the quiescent state up the rcu_node
hierarchy, fixing the bug. This ends up requiring a separate
lock_class_key per level of the rcu_node hierarchy, which this
patch also provides.
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: <12589088301770-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-22 17:53:48 +01:00
|
|
|
}
|
|
|
|
|
2010-11-30 06:56:39 +01:00
|
|
|
/*
|
|
|
|
* Advance a ->blkd_tasks-list pointer to the next entry, instead
|
|
|
|
* returning NULL if at the end of the list.
|
|
|
|
*/
|
|
|
|
static struct list_head *rcu_next_node_entry(struct task_struct *t,
|
|
|
|
struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
struct list_head *np;
|
|
|
|
|
|
|
|
np = t->rcu_node_entry.next;
|
|
|
|
if (np == &rnp->blkd_tasks)
|
|
|
|
np = NULL;
|
|
|
|
return np;
|
|
|
|
}
|
|
|
|
|
rcu: Fix grace-period-stall bug on large systems with CPU hotplug
When the last CPU of a given leaf rcu_node structure goes
offline, all of the tasks queued on that leaf rcu_node structure
(due to having blocked in their current RCU read-side critical
sections) are requeued onto the root rcu_node structure. This
requeuing is carried out by rcu_preempt_offline_tasks().
However, it is possible that these queued tasks are the only
thing preventing the leaf rcu_node structure from reporting a
quiescent state up the rcu_node hierarchy. Unfortunately, the
old code would fail to do this reporting, resulting in a
grace-period stall given the following sequence of events:
1. Kernel built for more than 32 CPUs on 32-bit systems or for more
than 64 CPUs on 64-bit systems, so that there is more than one
rcu_node structure. (Or CONFIG_RCU_FANOUT is artificially set
to a number smaller than CONFIG_NR_CPUS.)
2. The kernel is built with CONFIG_TREE_PREEMPT_RCU.
3. A task running on a CPU associated with a given leaf rcu_node
structure blocks while in an RCU read-side critical section
-and- that CPU has not yet passed through a quiescent state
for the current RCU grace period. This will cause the task
to be queued on the leaf rcu_node's blocked_tasks[] array, in
particular, on the element of this array corresponding to the
current grace period.
4. Each of the remaining CPUs corresponding to this same leaf rcu_node
structure pass through a quiescent state. However, the task is
still in its RCU read-side critical section, so these quiescent
states cannot be reported further up the rcu_node hierarchy.
Nevertheless, all bits in the leaf rcu_node structure's ->qsmask
field are now zero.
5. Each of the remaining CPUs go offline. (The events in step
#4 and #5 can happen in any order as long as each CPU passes
through a quiescent state before going offline.)
6. When the last CPU goes offline, __rcu_offline_cpu() will invoke
rcu_preempt_offline_tasks(), which will move the task to the
root rcu_node structure, but without reporting a quiescent state
up the rcu_node hierarchy (and this failure to report a quiescent
state is the bug).
But because this leaf rcu_node structure's ->qsmask field is
already zero and its ->block_tasks[] entries are all empty,
force_quiescent_state() will skip this rcu_node structure.
Therefore, grace periods are now hung.
This patch abstracts some code out of rcu_read_unlock_special(),
calling the result task_quiet() by analogy with cpu_quiet(), and
invokes task_quiet() from both rcu_read_lock_special() and
__rcu_offline_cpu(). Invoking task_quiet() from
__rcu_offline_cpu() reports the quiescent state up the rcu_node
hierarchy, fixing the bug. This ends up requiring a separate
lock_class_key per level of the rcu_node hierarchy, which this
patch also provides.
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: <12589088301770-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-22 17:53:48 +01:00
|
|
|
/*
|
|
|
|
* Handle special cases during rcu_read_unlock(), such as needing to
|
|
|
|
* notify RCU core processing or task having blocked during the RCU
|
|
|
|
* read-side critical section.
|
|
|
|
*/
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
static void rcu_read_unlock_special(struct task_struct *t)
|
|
|
|
{
|
|
|
|
int empty;
|
2009-12-02 21:10:15 +01:00
|
|
|
int empty_exp;
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
unsigned long flags;
|
2010-11-30 06:56:39 +01:00
|
|
|
struct list_head *np;
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
struct rcu_node *rnp;
|
|
|
|
int special;
|
|
|
|
|
|
|
|
/* NMI handlers cannot block and cannot safely manipulate state. */
|
|
|
|
if (in_nmi())
|
|
|
|
return;
|
|
|
|
|
|
|
|
local_irq_save(flags);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If RCU core is waiting for this CPU to exit critical section,
|
|
|
|
* let it know that we have done so.
|
|
|
|
*/
|
|
|
|
special = t->rcu_read_unlock_special;
|
|
|
|
if (special & RCU_READ_UNLOCK_NEED_QS) {
|
2009-09-13 18:15:10 +02:00
|
|
|
rcu_preempt_qs(smp_processor_id());
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/* Hardware IRQ handlers cannot block. */
|
|
|
|
if (in_irq()) {
|
|
|
|
local_irq_restore(flags);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Clean up if blocked during RCU read-side critical section. */
|
|
|
|
if (special & RCU_READ_UNLOCK_BLOCKED) {
|
|
|
|
t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_BLOCKED;
|
|
|
|
|
2009-08-27 23:58:16 +02:00
|
|
|
/*
|
|
|
|
* Remove this task from the list it blocked on. The
|
|
|
|
* task can migrate while we acquire the lock, but at
|
|
|
|
* most one time. So at most two passes through loop.
|
|
|
|
*/
|
|
|
|
for (;;) {
|
2009-08-28 00:00:12 +02:00
|
|
|
rnp = t->rcu_blocked_node;
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_lock(&rnp->lock); /* irqs already disabled. */
|
2009-08-28 00:00:12 +02:00
|
|
|
if (rnp == t->rcu_blocked_node)
|
2009-08-27 23:58:16 +02:00
|
|
|
break;
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */
|
2009-08-27 23:58:16 +02:00
|
|
|
}
|
2011-02-07 21:47:15 +01:00
|
|
|
empty = !rcu_preempt_blocked_readers_cgp(rnp);
|
2009-12-02 21:10:15 +01:00
|
|
|
empty_exp = !rcu_preempted_readers_exp(rnp);
|
|
|
|
smp_mb(); /* ensure expedited fastpath sees end of RCU c-s. */
|
2010-11-30 06:56:39 +01:00
|
|
|
np = rcu_next_node_entry(t, rnp);
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
list_del_init(&t->rcu_node_entry);
|
2010-11-30 06:56:39 +01:00
|
|
|
if (&t->rcu_node_entry == rnp->gp_tasks)
|
|
|
|
rnp->gp_tasks = np;
|
|
|
|
if (&t->rcu_node_entry == rnp->exp_tasks)
|
|
|
|
rnp->exp_tasks = np;
|
2011-02-07 21:47:15 +01:00
|
|
|
#ifdef CONFIG_RCU_BOOST
|
|
|
|
if (&t->rcu_node_entry == rnp->boost_tasks)
|
|
|
|
rnp->boost_tasks = np;
|
|
|
|
#endif /* #ifdef CONFIG_RCU_BOOST */
|
2009-08-27 23:58:16 +02:00
|
|
|
t->rcu_blocked_node = NULL;
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* If this was the last task on the current list, and if
|
|
|
|
* we aren't waiting on any CPUs, report the quiescent state.
|
2009-12-02 21:10:13 +01:00
|
|
|
* Note that rcu_report_unblock_qs_rnp() releases rnp->lock.
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
*/
|
rcu: Fix grace-period-stall bug on large systems with CPU hotplug
When the last CPU of a given leaf rcu_node structure goes
offline, all of the tasks queued on that leaf rcu_node structure
(due to having blocked in their current RCU read-side critical
sections) are requeued onto the root rcu_node structure. This
requeuing is carried out by rcu_preempt_offline_tasks().
However, it is possible that these queued tasks are the only
thing preventing the leaf rcu_node structure from reporting a
quiescent state up the rcu_node hierarchy. Unfortunately, the
old code would fail to do this reporting, resulting in a
grace-period stall given the following sequence of events:
1. Kernel built for more than 32 CPUs on 32-bit systems or for more
than 64 CPUs on 64-bit systems, so that there is more than one
rcu_node structure. (Or CONFIG_RCU_FANOUT is artificially set
to a number smaller than CONFIG_NR_CPUS.)
2. The kernel is built with CONFIG_TREE_PREEMPT_RCU.
3. A task running on a CPU associated with a given leaf rcu_node
structure blocks while in an RCU read-side critical section
-and- that CPU has not yet passed through a quiescent state
for the current RCU grace period. This will cause the task
to be queued on the leaf rcu_node's blocked_tasks[] array, in
particular, on the element of this array corresponding to the
current grace period.
4. Each of the remaining CPUs corresponding to this same leaf rcu_node
structure pass through a quiescent state. However, the task is
still in its RCU read-side critical section, so these quiescent
states cannot be reported further up the rcu_node hierarchy.
Nevertheless, all bits in the leaf rcu_node structure's ->qsmask
field are now zero.
5. Each of the remaining CPUs go offline. (The events in step
#4 and #5 can happen in any order as long as each CPU passes
through a quiescent state before going offline.)
6. When the last CPU goes offline, __rcu_offline_cpu() will invoke
rcu_preempt_offline_tasks(), which will move the task to the
root rcu_node structure, but without reporting a quiescent state
up the rcu_node hierarchy (and this failure to report a quiescent
state is the bug).
But because this leaf rcu_node structure's ->qsmask field is
already zero and its ->block_tasks[] entries are all empty,
force_quiescent_state() will skip this rcu_node structure.
Therefore, grace periods are now hung.
This patch abstracts some code out of rcu_read_unlock_special(),
calling the result task_quiet() by analogy with cpu_quiet(), and
invokes task_quiet() from both rcu_read_lock_special() and
__rcu_offline_cpu(). Invoking task_quiet() from
__rcu_offline_cpu() reports the quiescent state up the rcu_node
hierarchy, fixing the bug. This ends up requiring a separate
lock_class_key per level of the rcu_node hierarchy, which this
patch also provides.
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: <12589088301770-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-22 17:53:48 +01:00
|
|
|
if (empty)
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_unlock_irqrestore(&rnp->lock, flags);
|
rcu: Fix grace-period-stall bug on large systems with CPU hotplug
When the last CPU of a given leaf rcu_node structure goes
offline, all of the tasks queued on that leaf rcu_node structure
(due to having blocked in their current RCU read-side critical
sections) are requeued onto the root rcu_node structure. This
requeuing is carried out by rcu_preempt_offline_tasks().
However, it is possible that these queued tasks are the only
thing preventing the leaf rcu_node structure from reporting a
quiescent state up the rcu_node hierarchy. Unfortunately, the
old code would fail to do this reporting, resulting in a
grace-period stall given the following sequence of events:
1. Kernel built for more than 32 CPUs on 32-bit systems or for more
than 64 CPUs on 64-bit systems, so that there is more than one
rcu_node structure. (Or CONFIG_RCU_FANOUT is artificially set
to a number smaller than CONFIG_NR_CPUS.)
2. The kernel is built with CONFIG_TREE_PREEMPT_RCU.
3. A task running on a CPU associated with a given leaf rcu_node
structure blocks while in an RCU read-side critical section
-and- that CPU has not yet passed through a quiescent state
for the current RCU grace period. This will cause the task
to be queued on the leaf rcu_node's blocked_tasks[] array, in
particular, on the element of this array corresponding to the
current grace period.
4. Each of the remaining CPUs corresponding to this same leaf rcu_node
structure pass through a quiescent state. However, the task is
still in its RCU read-side critical section, so these quiescent
states cannot be reported further up the rcu_node hierarchy.
Nevertheless, all bits in the leaf rcu_node structure's ->qsmask
field are now zero.
5. Each of the remaining CPUs go offline. (The events in step
#4 and #5 can happen in any order as long as each CPU passes
through a quiescent state before going offline.)
6. When the last CPU goes offline, __rcu_offline_cpu() will invoke
rcu_preempt_offline_tasks(), which will move the task to the
root rcu_node structure, but without reporting a quiescent state
up the rcu_node hierarchy (and this failure to report a quiescent
state is the bug).
But because this leaf rcu_node structure's ->qsmask field is
already zero and its ->block_tasks[] entries are all empty,
force_quiescent_state() will skip this rcu_node structure.
Therefore, grace periods are now hung.
This patch abstracts some code out of rcu_read_unlock_special(),
calling the result task_quiet() by analogy with cpu_quiet(), and
invokes task_quiet() from both rcu_read_lock_special() and
__rcu_offline_cpu(). Invoking task_quiet() from
__rcu_offline_cpu() reports the quiescent state up the rcu_node
hierarchy, fixing the bug. This ends up requiring a separate
lock_class_key per level of the rcu_node hierarchy, which this
patch also provides.
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: <12589088301770-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-22 17:53:48 +01:00
|
|
|
else
|
2009-12-02 21:10:13 +01:00
|
|
|
rcu_report_unblock_qs_rnp(rnp, flags);
|
2009-12-02 21:10:15 +01:00
|
|
|
|
2011-02-07 21:47:15 +01:00
|
|
|
#ifdef CONFIG_RCU_BOOST
|
|
|
|
/* Unboost if we were boosted. */
|
|
|
|
if (special & RCU_READ_UNLOCK_BOOSTED) {
|
|
|
|
t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_BOOSTED;
|
|
|
|
rt_mutex_unlock(t->rcu_boost_mutex);
|
|
|
|
t->rcu_boost_mutex = NULL;
|
|
|
|
}
|
|
|
|
#endif /* #ifdef CONFIG_RCU_BOOST */
|
|
|
|
|
2009-12-02 21:10:15 +01:00
|
|
|
/*
|
|
|
|
* If this was the last task on the expedited lists,
|
|
|
|
* then we need to report up the rcu_node hierarchy.
|
|
|
|
*/
|
|
|
|
if (!empty_exp && !rcu_preempted_readers_exp(rnp))
|
|
|
|
rcu_report_exp_rnp(&rcu_preempt_state, rnp);
|
rcu: Fix grace-period-stall bug on large systems with CPU hotplug
When the last CPU of a given leaf rcu_node structure goes
offline, all of the tasks queued on that leaf rcu_node structure
(due to having blocked in their current RCU read-side critical
sections) are requeued onto the root rcu_node structure. This
requeuing is carried out by rcu_preempt_offline_tasks().
However, it is possible that these queued tasks are the only
thing preventing the leaf rcu_node structure from reporting a
quiescent state up the rcu_node hierarchy. Unfortunately, the
old code would fail to do this reporting, resulting in a
grace-period stall given the following sequence of events:
1. Kernel built for more than 32 CPUs on 32-bit systems or for more
than 64 CPUs on 64-bit systems, so that there is more than one
rcu_node structure. (Or CONFIG_RCU_FANOUT is artificially set
to a number smaller than CONFIG_NR_CPUS.)
2. The kernel is built with CONFIG_TREE_PREEMPT_RCU.
3. A task running on a CPU associated with a given leaf rcu_node
structure blocks while in an RCU read-side critical section
-and- that CPU has not yet passed through a quiescent state
for the current RCU grace period. This will cause the task
to be queued on the leaf rcu_node's blocked_tasks[] array, in
particular, on the element of this array corresponding to the
current grace period.
4. Each of the remaining CPUs corresponding to this same leaf rcu_node
structure pass through a quiescent state. However, the task is
still in its RCU read-side critical section, so these quiescent
states cannot be reported further up the rcu_node hierarchy.
Nevertheless, all bits in the leaf rcu_node structure's ->qsmask
field are now zero.
5. Each of the remaining CPUs go offline. (The events in step
#4 and #5 can happen in any order as long as each CPU passes
through a quiescent state before going offline.)
6. When the last CPU goes offline, __rcu_offline_cpu() will invoke
rcu_preempt_offline_tasks(), which will move the task to the
root rcu_node structure, but without reporting a quiescent state
up the rcu_node hierarchy (and this failure to report a quiescent
state is the bug).
But because this leaf rcu_node structure's ->qsmask field is
already zero and its ->block_tasks[] entries are all empty,
force_quiescent_state() will skip this rcu_node structure.
Therefore, grace periods are now hung.
This patch abstracts some code out of rcu_read_unlock_special(),
calling the result task_quiet() by analogy with cpu_quiet(), and
invokes task_quiet() from both rcu_read_lock_special() and
__rcu_offline_cpu(). Invoking task_quiet() from
__rcu_offline_cpu() reports the quiescent state up the rcu_node
hierarchy, fixing the bug. This ends up requiring a separate
lock_class_key per level of the rcu_node hierarchy, which this
patch also provides.
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: <12589088301770-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-22 17:53:48 +01:00
|
|
|
} else {
|
|
|
|
local_irq_restore(flags);
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Tree-preemptible RCU implementation for rcu_read_unlock().
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
* Decrement ->rcu_read_lock_nesting. If the result is zero (outermost
|
|
|
|
* rcu_read_unlock()) and ->rcu_read_unlock_special is non-zero, then
|
|
|
|
* invoke rcu_read_unlock_special() to clean up after a context switch
|
|
|
|
* in an RCU read-side critical section and other special cases.
|
|
|
|
*/
|
|
|
|
void __rcu_read_unlock(void)
|
|
|
|
{
|
|
|
|
struct task_struct *t = current;
|
|
|
|
|
|
|
|
barrier(); /* needed if we ever invoke rcu_read_unlock in rcutree.c */
|
2010-08-20 01:57:45 +02:00
|
|
|
--t->rcu_read_lock_nesting;
|
|
|
|
barrier(); /* decrement before load of ->rcu_read_unlock_special */
|
|
|
|
if (t->rcu_read_lock_nesting == 0 &&
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
unlikely(ACCESS_ONCE(t->rcu_read_unlock_special)))
|
|
|
|
rcu_read_unlock_special(t);
|
2010-01-05 01:04:01 +01:00
|
|
|
#ifdef CONFIG_PROVE_LOCKING
|
|
|
|
WARN_ON_ONCE(ACCESS_ONCE(t->rcu_read_lock_nesting) < 0);
|
|
|
|
#endif /* #ifdef CONFIG_PROVE_LOCKING */
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(__rcu_read_unlock);
|
|
|
|
|
2010-02-23 02:05:05 +01:00
|
|
|
#ifdef CONFIG_RCU_CPU_STALL_VERBOSE
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Dump detailed information for all tasks blocking the current RCU
|
|
|
|
* grace period on the specified rcu_node structure.
|
|
|
|
*/
|
|
|
|
static void rcu_print_detail_task_stall_rnp(struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
struct task_struct *t;
|
|
|
|
|
2011-02-07 21:47:15 +01:00
|
|
|
if (!rcu_preempt_blocked_readers_cgp(rnp))
|
2010-11-30 06:56:39 +01:00
|
|
|
return;
|
|
|
|
raw_spin_lock_irqsave(&rnp->lock, flags);
|
|
|
|
t = list_entry(rnp->gp_tasks,
|
|
|
|
struct task_struct, rcu_node_entry);
|
|
|
|
list_for_each_entry_continue(t, &rnp->blkd_tasks, rcu_node_entry)
|
|
|
|
sched_show_task(t);
|
|
|
|
raw_spin_unlock_irqrestore(&rnp->lock, flags);
|
2010-02-23 02:05:05 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Dump detailed information for all tasks blocking the current RCU
|
|
|
|
* grace period.
|
|
|
|
*/
|
|
|
|
static void rcu_print_detail_task_stall(struct rcu_state *rsp)
|
|
|
|
{
|
|
|
|
struct rcu_node *rnp = rcu_get_root(rsp);
|
|
|
|
|
|
|
|
rcu_print_detail_task_stall_rnp(rnp);
|
|
|
|
rcu_for_each_leaf_node(rsp, rnp)
|
|
|
|
rcu_print_detail_task_stall_rnp(rnp);
|
|
|
|
}
|
|
|
|
|
|
|
|
#else /* #ifdef CONFIG_RCU_CPU_STALL_VERBOSE */
|
|
|
|
|
|
|
|
static void rcu_print_detail_task_stall(struct rcu_state *rsp)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
#endif /* #else #ifdef CONFIG_RCU_CPU_STALL_VERBOSE */
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
/*
|
|
|
|
* Scan the current list of tasks blocked within RCU read-side critical
|
|
|
|
* sections, printing out the tid of each.
|
|
|
|
*/
|
|
|
|
static void rcu_print_task_stall(struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
struct task_struct *t;
|
|
|
|
|
2011-02-07 21:47:15 +01:00
|
|
|
if (!rcu_preempt_blocked_readers_cgp(rnp))
|
2010-11-30 06:56:39 +01:00
|
|
|
return;
|
|
|
|
t = list_entry(rnp->gp_tasks,
|
|
|
|
struct task_struct, rcu_node_entry);
|
|
|
|
list_for_each_entry_continue(t, &rnp->blkd_tasks, rcu_node_entry)
|
|
|
|
printk(" P%d", t->pid);
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
}
|
|
|
|
|
2010-08-10 23:28:53 +02:00
|
|
|
/*
|
|
|
|
* Suppress preemptible RCU's CPU stall warnings by pushing the
|
|
|
|
* time of the next stall-warning message comfortably far into the
|
|
|
|
* future.
|
|
|
|
*/
|
|
|
|
static void rcu_preempt_stall_reset(void)
|
|
|
|
{
|
|
|
|
rcu_preempt_state.jiffies_stall = jiffies + ULONG_MAX / 2;
|
|
|
|
}
|
|
|
|
|
2009-09-13 18:15:09 +02:00
|
|
|
/*
|
|
|
|
* Check that the list of blocked tasks for the newly completed grace
|
|
|
|
* period is in fact empty. It is a serious bug to complete a grace
|
|
|
|
* period that still has RCU readers blocked! This function must be
|
|
|
|
* invoked -before- updating this rnp's ->gpnum, and the rnp's ->lock
|
|
|
|
* must be held by the caller.
|
2010-11-30 06:56:39 +01:00
|
|
|
*
|
|
|
|
* Also, if there are blocked tasks on the list, they automatically
|
|
|
|
* block the newly created grace period, so set up ->gp_tasks accordingly.
|
2009-09-13 18:15:09 +02:00
|
|
|
*/
|
|
|
|
static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp)
|
|
|
|
{
|
2011-02-07 21:47:15 +01:00
|
|
|
WARN_ON_ONCE(rcu_preempt_blocked_readers_cgp(rnp));
|
2010-11-30 06:56:39 +01:00
|
|
|
if (!list_empty(&rnp->blkd_tasks))
|
|
|
|
rnp->gp_tasks = rnp->blkd_tasks.next;
|
2009-09-18 18:50:17 +02:00
|
|
|
WARN_ON_ONCE(rnp->qsmask);
|
2009-09-13 18:15:09 +02:00
|
|
|
}
|
|
|
|
|
2009-08-24 18:42:01 +02:00
|
|
|
#ifdef CONFIG_HOTPLUG_CPU
|
|
|
|
|
2009-08-27 23:58:16 +02:00
|
|
|
/*
|
|
|
|
* Handle tasklist migration for case in which all CPUs covered by the
|
|
|
|
* specified rcu_node have gone offline. Move them up to the root
|
|
|
|
* rcu_node. The reason for not just moving them to the immediate
|
|
|
|
* parent is to remove the need for rcu_read_unlock_special() to
|
|
|
|
* make more than two attempts to acquire the target rcu_node's lock.
|
rcu: Fix grace-period-stall bug on large systems with CPU hotplug
When the last CPU of a given leaf rcu_node structure goes
offline, all of the tasks queued on that leaf rcu_node structure
(due to having blocked in their current RCU read-side critical
sections) are requeued onto the root rcu_node structure. This
requeuing is carried out by rcu_preempt_offline_tasks().
However, it is possible that these queued tasks are the only
thing preventing the leaf rcu_node structure from reporting a
quiescent state up the rcu_node hierarchy. Unfortunately, the
old code would fail to do this reporting, resulting in a
grace-period stall given the following sequence of events:
1. Kernel built for more than 32 CPUs on 32-bit systems or for more
than 64 CPUs on 64-bit systems, so that there is more than one
rcu_node structure. (Or CONFIG_RCU_FANOUT is artificially set
to a number smaller than CONFIG_NR_CPUS.)
2. The kernel is built with CONFIG_TREE_PREEMPT_RCU.
3. A task running on a CPU associated with a given leaf rcu_node
structure blocks while in an RCU read-side critical section
-and- that CPU has not yet passed through a quiescent state
for the current RCU grace period. This will cause the task
to be queued on the leaf rcu_node's blocked_tasks[] array, in
particular, on the element of this array corresponding to the
current grace period.
4. Each of the remaining CPUs corresponding to this same leaf rcu_node
structure pass through a quiescent state. However, the task is
still in its RCU read-side critical section, so these quiescent
states cannot be reported further up the rcu_node hierarchy.
Nevertheless, all bits in the leaf rcu_node structure's ->qsmask
field are now zero.
5. Each of the remaining CPUs go offline. (The events in step
#4 and #5 can happen in any order as long as each CPU passes
through a quiescent state before going offline.)
6. When the last CPU goes offline, __rcu_offline_cpu() will invoke
rcu_preempt_offline_tasks(), which will move the task to the
root rcu_node structure, but without reporting a quiescent state
up the rcu_node hierarchy (and this failure to report a quiescent
state is the bug).
But because this leaf rcu_node structure's ->qsmask field is
already zero and its ->block_tasks[] entries are all empty,
force_quiescent_state() will skip this rcu_node structure.
Therefore, grace periods are now hung.
This patch abstracts some code out of rcu_read_unlock_special(),
calling the result task_quiet() by analogy with cpu_quiet(), and
invokes task_quiet() from both rcu_read_lock_special() and
__rcu_offline_cpu(). Invoking task_quiet() from
__rcu_offline_cpu() reports the quiescent state up the rcu_node
hierarchy, fixing the bug. This ends up requiring a separate
lock_class_key per level of the rcu_node hierarchy, which this
patch also provides.
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: <12589088301770-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-22 17:53:48 +01:00
|
|
|
* Returns true if there were tasks blocking the current RCU grace
|
|
|
|
* period.
|
2009-08-27 23:58:16 +02:00
|
|
|
*
|
rcu: Fix TREE_PREEMPT_RCU CPU_HOTPLUG bad-luck hang
If the following sequence of events occurs, then
TREE_PREEMPT_RCU will hang waiting for a grace period to
complete, eventually OOMing the system:
o A TREE_PREEMPT_RCU build of the kernel is booted on a system
with more than 64 physical CPUs present (32 on a 32-bit system).
Alternatively, a TREE_PREEMPT_RCU build of the kernel is booted
with RCU_FANOUT set to a sufficiently small value that the
physical CPUs populate two or more leaf rcu_node structures.
o A task is preempted in an RCU read-side critical section
while running on a CPU corresponding to a given leaf rcu_node
structure.
o All CPUs corresponding to this same leaf rcu_node structure
record quiescent states for the current grace period.
o All of these same CPUs go offline (hence the need for enough
physical CPUs to populate more than one leaf rcu_node structure).
This causes the preempted task to be moved to the root rcu_node
structure.
At this point, there is nothing left to cause the quiescent
state to be propagated up the rcu_node tree, so the current
grace period never completes.
The simplest fix, especially after considering the deadlock
possibilities, is to detect this situation when the last CPU is
offlined, and to set that CPU's ->qsmask bit in its leaf
rcu_node structure. This will cause the next invocation of
force_quiescent_state() to end the grace period.
Without this fix, this hang can be triggered in an hour or so on
some machines with rcutorture and random CPU onlining/offlining.
With this fix, these same machines pass a full 10 hours of this
sort of abuse.
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: <20091015162614.GA19131@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-15 18:26:14 +02:00
|
|
|
* Returns 1 if there was previously a task blocking the current grace
|
|
|
|
* period on the specified rcu_node structure.
|
|
|
|
*
|
2009-08-27 23:58:16 +02:00
|
|
|
* The caller must hold rnp->lock with irqs disabled.
|
|
|
|
*/
|
rcu: Fix TREE_PREEMPT_RCU CPU_HOTPLUG bad-luck hang
If the following sequence of events occurs, then
TREE_PREEMPT_RCU will hang waiting for a grace period to
complete, eventually OOMing the system:
o A TREE_PREEMPT_RCU build of the kernel is booted on a system
with more than 64 physical CPUs present (32 on a 32-bit system).
Alternatively, a TREE_PREEMPT_RCU build of the kernel is booted
with RCU_FANOUT set to a sufficiently small value that the
physical CPUs populate two or more leaf rcu_node structures.
o A task is preempted in an RCU read-side critical section
while running on a CPU corresponding to a given leaf rcu_node
structure.
o All CPUs corresponding to this same leaf rcu_node structure
record quiescent states for the current grace period.
o All of these same CPUs go offline (hence the need for enough
physical CPUs to populate more than one leaf rcu_node structure).
This causes the preempted task to be moved to the root rcu_node
structure.
At this point, there is nothing left to cause the quiescent
state to be propagated up the rcu_node tree, so the current
grace period never completes.
The simplest fix, especially after considering the deadlock
possibilities, is to detect this situation when the last CPU is
offlined, and to set that CPU's ->qsmask bit in its leaf
rcu_node structure. This will cause the next invocation of
force_quiescent_state() to end the grace period.
Without this fix, this hang can be triggered in an hour or so on
some machines with rcutorture and random CPU onlining/offlining.
With this fix, these same machines pass a full 10 hours of this
sort of abuse.
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: <20091015162614.GA19131@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-15 18:26:14 +02:00
|
|
|
static int rcu_preempt_offline_tasks(struct rcu_state *rsp,
|
|
|
|
struct rcu_node *rnp,
|
|
|
|
struct rcu_data *rdp)
|
2009-08-27 23:58:16 +02:00
|
|
|
{
|
|
|
|
struct list_head *lp;
|
|
|
|
struct list_head *lp_root;
|
2009-12-02 21:10:15 +01:00
|
|
|
int retval = 0;
|
2009-08-27 23:58:16 +02:00
|
|
|
struct rcu_node *rnp_root = rcu_get_root(rsp);
|
2010-11-30 06:56:39 +01:00
|
|
|
struct task_struct *t;
|
2009-08-27 23:58:16 +02:00
|
|
|
|
2009-08-28 00:00:12 +02:00
|
|
|
if (rnp == rnp_root) {
|
|
|
|
WARN_ONCE(1, "Last CPU thought to be offlined?");
|
rcu: Fix TREE_PREEMPT_RCU CPU_HOTPLUG bad-luck hang
If the following sequence of events occurs, then
TREE_PREEMPT_RCU will hang waiting for a grace period to
complete, eventually OOMing the system:
o A TREE_PREEMPT_RCU build of the kernel is booted on a system
with more than 64 physical CPUs present (32 on a 32-bit system).
Alternatively, a TREE_PREEMPT_RCU build of the kernel is booted
with RCU_FANOUT set to a sufficiently small value that the
physical CPUs populate two or more leaf rcu_node structures.
o A task is preempted in an RCU read-side critical section
while running on a CPU corresponding to a given leaf rcu_node
structure.
o All CPUs corresponding to this same leaf rcu_node structure
record quiescent states for the current grace period.
o All of these same CPUs go offline (hence the need for enough
physical CPUs to populate more than one leaf rcu_node structure).
This causes the preempted task to be moved to the root rcu_node
structure.
At this point, there is nothing left to cause the quiescent
state to be propagated up the rcu_node tree, so the current
grace period never completes.
The simplest fix, especially after considering the deadlock
possibilities, is to detect this situation when the last CPU is
offlined, and to set that CPU's ->qsmask bit in its leaf
rcu_node structure. This will cause the next invocation of
force_quiescent_state() to end the grace period.
Without this fix, this hang can be triggered in an hour or so on
some machines with rcutorture and random CPU onlining/offlining.
With this fix, these same machines pass a full 10 hours of this
sort of abuse.
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: <20091015162614.GA19131@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-15 18:26:14 +02:00
|
|
|
return 0; /* Shouldn't happen: at least one CPU online. */
|
2009-08-28 00:00:12 +02:00
|
|
|
}
|
2010-11-30 06:56:39 +01:00
|
|
|
|
|
|
|
/* If we are on an internal node, complain bitterly. */
|
|
|
|
WARN_ON_ONCE(rnp != rdp->mynode);
|
2009-08-27 23:58:16 +02:00
|
|
|
|
|
|
|
/*
|
2010-11-30 06:56:39 +01:00
|
|
|
* Move tasks up to root rcu_node. Don't try to get fancy for
|
|
|
|
* this corner-case operation -- just put this node's tasks
|
|
|
|
* at the head of the root node's list, and update the root node's
|
|
|
|
* ->gp_tasks and ->exp_tasks pointers to those of this node's,
|
|
|
|
* if non-NULL. This might result in waiting for more tasks than
|
|
|
|
* absolutely necessary, but this is a good performance/complexity
|
|
|
|
* tradeoff.
|
2009-08-27 23:58:16 +02:00
|
|
|
*/
|
2011-02-07 21:47:15 +01:00
|
|
|
if (rcu_preempt_blocked_readers_cgp(rnp))
|
2009-12-02 21:10:15 +01:00
|
|
|
retval |= RCU_OFL_TASKS_NORM_GP;
|
|
|
|
if (rcu_preempted_readers_exp(rnp))
|
|
|
|
retval |= RCU_OFL_TASKS_EXP_GP;
|
2010-11-30 06:56:39 +01:00
|
|
|
lp = &rnp->blkd_tasks;
|
|
|
|
lp_root = &rnp_root->blkd_tasks;
|
|
|
|
while (!list_empty(lp)) {
|
|
|
|
t = list_entry(lp->next, typeof(*t), rcu_node_entry);
|
|
|
|
raw_spin_lock(&rnp_root->lock); /* irqs already disabled */
|
|
|
|
list_del(&t->rcu_node_entry);
|
|
|
|
t->rcu_blocked_node = rnp_root;
|
|
|
|
list_add(&t->rcu_node_entry, lp_root);
|
|
|
|
if (&t->rcu_node_entry == rnp->gp_tasks)
|
|
|
|
rnp_root->gp_tasks = rnp->gp_tasks;
|
|
|
|
if (&t->rcu_node_entry == rnp->exp_tasks)
|
|
|
|
rnp_root->exp_tasks = rnp->exp_tasks;
|
2011-02-07 21:47:15 +01:00
|
|
|
#ifdef CONFIG_RCU_BOOST
|
|
|
|
if (&t->rcu_node_entry == rnp->boost_tasks)
|
|
|
|
rnp_root->boost_tasks = rnp->boost_tasks;
|
|
|
|
#endif /* #ifdef CONFIG_RCU_BOOST */
|
2010-11-30 06:56:39 +01:00
|
|
|
raw_spin_unlock(&rnp_root->lock); /* irqs still disabled */
|
2009-08-27 23:58:16 +02:00
|
|
|
}
|
2011-02-07 21:47:15 +01:00
|
|
|
|
|
|
|
#ifdef CONFIG_RCU_BOOST
|
|
|
|
/* In case root is being boosted and leaf is not. */
|
|
|
|
raw_spin_lock(&rnp_root->lock); /* irqs already disabled */
|
|
|
|
if (rnp_root->boost_tasks != NULL &&
|
|
|
|
rnp_root->boost_tasks != rnp_root->gp_tasks)
|
|
|
|
rnp_root->boost_tasks = rnp_root->gp_tasks;
|
|
|
|
raw_spin_unlock(&rnp_root->lock); /* irqs still disabled */
|
|
|
|
#endif /* #ifdef CONFIG_RCU_BOOST */
|
|
|
|
|
2010-11-30 06:56:39 +01:00
|
|
|
rnp->gp_tasks = NULL;
|
|
|
|
rnp->exp_tasks = NULL;
|
rcu: Fix TREE_PREEMPT_RCU CPU_HOTPLUG bad-luck hang
If the following sequence of events occurs, then
TREE_PREEMPT_RCU will hang waiting for a grace period to
complete, eventually OOMing the system:
o A TREE_PREEMPT_RCU build of the kernel is booted on a system
with more than 64 physical CPUs present (32 on a 32-bit system).
Alternatively, a TREE_PREEMPT_RCU build of the kernel is booted
with RCU_FANOUT set to a sufficiently small value that the
physical CPUs populate two or more leaf rcu_node structures.
o A task is preempted in an RCU read-side critical section
while running on a CPU corresponding to a given leaf rcu_node
structure.
o All CPUs corresponding to this same leaf rcu_node structure
record quiescent states for the current grace period.
o All of these same CPUs go offline (hence the need for enough
physical CPUs to populate more than one leaf rcu_node structure).
This causes the preempted task to be moved to the root rcu_node
structure.
At this point, there is nothing left to cause the quiescent
state to be propagated up the rcu_node tree, so the current
grace period never completes.
The simplest fix, especially after considering the deadlock
possibilities, is to detect this situation when the last CPU is
offlined, and to set that CPU's ->qsmask bit in its leaf
rcu_node structure. This will cause the next invocation of
force_quiescent_state() to end the grace period.
Without this fix, this hang can be triggered in an hour or so on
some machines with rcutorture and random CPU onlining/offlining.
With this fix, these same machines pass a full 10 hours of this
sort of abuse.
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: <20091015162614.GA19131@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-15 18:26:14 +02:00
|
|
|
return retval;
|
2009-08-27 23:58:16 +02:00
|
|
|
}
|
|
|
|
|
2009-08-24 18:42:01 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Do CPU-offline processing for preemptible RCU.
|
2009-08-24 18:42:01 +02:00
|
|
|
*/
|
|
|
|
static void rcu_preempt_offline_cpu(int cpu)
|
|
|
|
{
|
|
|
|
__rcu_offline_cpu(cpu, &rcu_preempt_state);
|
|
|
|
}
|
|
|
|
|
|
|
|
#endif /* #ifdef CONFIG_HOTPLUG_CPU */
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
/*
|
|
|
|
* Check for a quiescent state from the current CPU. When a task blocks,
|
|
|
|
* the task is recorded in the corresponding CPU's rcu_node structure,
|
|
|
|
* which is checked elsewhere.
|
|
|
|
*
|
|
|
|
* Caller must disable hard irqs.
|
|
|
|
*/
|
|
|
|
static void rcu_preempt_check_callbacks(int cpu)
|
|
|
|
{
|
|
|
|
struct task_struct *t = current;
|
|
|
|
|
|
|
|
if (t->rcu_read_lock_nesting == 0) {
|
2009-09-13 18:15:10 +02:00
|
|
|
rcu_preempt_qs(cpu);
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
return;
|
|
|
|
}
|
2009-09-18 19:28:19 +02:00
|
|
|
if (per_cpu(rcu_preempt_data, cpu).qs_pending)
|
2009-09-13 18:15:10 +02:00
|
|
|
t->rcu_read_unlock_special |= RCU_READ_UNLOCK_NEED_QS;
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Process callbacks for preemptible RCU.
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
*/
|
|
|
|
static void rcu_preempt_process_callbacks(void)
|
|
|
|
{
|
|
|
|
__rcu_process_callbacks(&rcu_preempt_state,
|
|
|
|
&__get_cpu_var(rcu_preempt_data));
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Queue a preemptible-RCU callback for invocation after a grace period.
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
*/
|
|
|
|
void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
|
|
|
|
{
|
|
|
|
__call_rcu(head, func, &rcu_preempt_state);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(call_rcu);
|
|
|
|
|
2009-11-22 17:53:50 +01:00
|
|
|
/**
|
|
|
|
* synchronize_rcu - wait until a grace period has elapsed.
|
|
|
|
*
|
|
|
|
* Control will return to the caller some time after a full grace
|
|
|
|
* period has elapsed, in other words after all currently executing RCU
|
2010-07-09 02:38:59 +02:00
|
|
|
* read-side critical sections have completed. Note, however, that
|
|
|
|
* upon return from synchronize_rcu(), the caller might well be executing
|
|
|
|
* concurrently with new RCU read-side critical sections that began while
|
|
|
|
* synchronize_rcu() was waiting. RCU read-side critical sections are
|
|
|
|
* delimited by rcu_read_lock() and rcu_read_unlock(), and may be nested.
|
2009-11-22 17:53:50 +01:00
|
|
|
*/
|
|
|
|
void synchronize_rcu(void)
|
|
|
|
{
|
|
|
|
struct rcu_synchronize rcu;
|
|
|
|
|
|
|
|
if (!rcu_scheduler_active)
|
|
|
|
return;
|
|
|
|
|
2010-05-11 02:12:17 +02:00
|
|
|
init_rcu_head_on_stack(&rcu.head);
|
2009-11-22 17:53:50 +01:00
|
|
|
init_completion(&rcu.completion);
|
|
|
|
/* Will wake me after RCU finished. */
|
|
|
|
call_rcu(&rcu.head, wakeme_after_rcu);
|
|
|
|
/* Wait for it. */
|
|
|
|
wait_for_completion(&rcu.completion);
|
2010-05-11 02:12:17 +02:00
|
|
|
destroy_rcu_head_on_stack(&rcu.head);
|
2009-11-22 17:53:50 +01:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(synchronize_rcu);
|
|
|
|
|
2009-12-02 21:10:15 +01:00
|
|
|
static DECLARE_WAIT_QUEUE_HEAD(sync_rcu_preempt_exp_wq);
|
|
|
|
static long sync_rcu_preempt_exp_count;
|
|
|
|
static DEFINE_MUTEX(sync_rcu_preempt_exp_mutex);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Return non-zero if there are any tasks in RCU read-side critical
|
|
|
|
* sections blocking the current preemptible-RCU expedited grace period.
|
|
|
|
* If there is no preemptible-RCU expedited grace period currently in
|
|
|
|
* progress, returns zero unconditionally.
|
|
|
|
*/
|
|
|
|
static int rcu_preempted_readers_exp(struct rcu_node *rnp)
|
|
|
|
{
|
2010-11-30 06:56:39 +01:00
|
|
|
return rnp->exp_tasks != NULL;
|
2009-12-02 21:10:15 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* return non-zero if there is no RCU expedited grace period in progress
|
|
|
|
* for the specified rcu_node structure, in other words, if all CPUs and
|
|
|
|
* tasks covered by the specified rcu_node structure have done their bit
|
|
|
|
* for the current expedited grace period. Works only for preemptible
|
|
|
|
* RCU -- other RCU implementation use other means.
|
|
|
|
*
|
|
|
|
* Caller must hold sync_rcu_preempt_exp_mutex.
|
|
|
|
*/
|
|
|
|
static int sync_rcu_preempt_exp_done(struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
return !rcu_preempted_readers_exp(rnp) &&
|
|
|
|
ACCESS_ONCE(rnp->expmask) == 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Report the exit from RCU read-side critical section for the last task
|
|
|
|
* that queued itself during or before the current expedited preemptible-RCU
|
|
|
|
* grace period. This event is reported either to the rcu_node structure on
|
|
|
|
* which the task was queued or to one of that rcu_node structure's ancestors,
|
|
|
|
* recursively up the tree. (Calm down, calm down, we do the recursion
|
|
|
|
* iteratively!)
|
|
|
|
*
|
|
|
|
* Caller must hold sync_rcu_preempt_exp_mutex.
|
|
|
|
*/
|
|
|
|
static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
unsigned long mask;
|
|
|
|
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_lock_irqsave(&rnp->lock, flags);
|
2009-12-02 21:10:15 +01:00
|
|
|
for (;;) {
|
|
|
|
if (!sync_rcu_preempt_exp_done(rnp))
|
|
|
|
break;
|
|
|
|
if (rnp->parent == NULL) {
|
|
|
|
wake_up(&sync_rcu_preempt_exp_wq);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
mask = rnp->grpmask;
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_unlock(&rnp->lock); /* irqs remain disabled */
|
2009-12-02 21:10:15 +01:00
|
|
|
rnp = rnp->parent;
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_lock(&rnp->lock); /* irqs already disabled */
|
2009-12-02 21:10:15 +01:00
|
|
|
rnp->expmask &= ~mask;
|
|
|
|
}
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_unlock_irqrestore(&rnp->lock, flags);
|
2009-12-02 21:10:15 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Snapshot the tasks blocking the newly started preemptible-RCU expedited
|
|
|
|
* grace period for the specified rcu_node structure. If there are no such
|
|
|
|
* tasks, report it up the rcu_node hierarchy.
|
|
|
|
*
|
|
|
|
* Caller must hold sync_rcu_preempt_exp_mutex and rsp->onofflock.
|
|
|
|
*/
|
|
|
|
static void
|
|
|
|
sync_rcu_preempt_exp_init(struct rcu_state *rsp, struct rcu_node *rnp)
|
|
|
|
{
|
2011-05-05 06:43:49 +02:00
|
|
|
unsigned long flags;
|
2010-11-30 06:56:39 +01:00
|
|
|
int must_wait = 0;
|
2009-12-02 21:10:15 +01:00
|
|
|
|
2011-05-05 06:43:49 +02:00
|
|
|
raw_spin_lock_irqsave(&rnp->lock, flags);
|
|
|
|
if (list_empty(&rnp->blkd_tasks))
|
|
|
|
raw_spin_unlock_irqrestore(&rnp->lock, flags);
|
|
|
|
else {
|
2010-11-30 06:56:39 +01:00
|
|
|
rnp->exp_tasks = rnp->blkd_tasks.next;
|
2011-05-05 06:43:49 +02:00
|
|
|
rcu_initiate_boost(rnp, flags); /* releases rnp->lock */
|
2010-11-30 06:56:39 +01:00
|
|
|
must_wait = 1;
|
|
|
|
}
|
2009-12-02 21:10:15 +01:00
|
|
|
if (!must_wait)
|
|
|
|
rcu_report_exp_rnp(rsp, rnp);
|
|
|
|
}
|
|
|
|
|
2009-10-14 19:15:56 +02:00
|
|
|
/*
|
2009-12-02 21:10:15 +01:00
|
|
|
* Wait for an rcu-preempt grace period, but expedite it. The basic idea
|
|
|
|
* is to invoke synchronize_sched_expedited() to push all the tasks to
|
2010-11-30 06:56:39 +01:00
|
|
|
* the ->blkd_tasks lists and wait for this list to drain.
|
2009-10-14 19:15:56 +02:00
|
|
|
*/
|
|
|
|
void synchronize_rcu_expedited(void)
|
|
|
|
{
|
2009-12-02 21:10:15 +01:00
|
|
|
unsigned long flags;
|
|
|
|
struct rcu_node *rnp;
|
|
|
|
struct rcu_state *rsp = &rcu_preempt_state;
|
|
|
|
long snap;
|
|
|
|
int trycount = 0;
|
|
|
|
|
|
|
|
smp_mb(); /* Caller's modifications seen first by other CPUs. */
|
|
|
|
snap = ACCESS_ONCE(sync_rcu_preempt_exp_count) + 1;
|
|
|
|
smp_mb(); /* Above access cannot bleed into critical section. */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Acquire lock, falling back to synchronize_rcu() if too many
|
|
|
|
* lock-acquisition failures. Of course, if someone does the
|
|
|
|
* expedited grace period for us, just leave.
|
|
|
|
*/
|
|
|
|
while (!mutex_trylock(&sync_rcu_preempt_exp_mutex)) {
|
|
|
|
if (trycount++ < 10)
|
|
|
|
udelay(trycount * num_online_cpus());
|
|
|
|
else {
|
|
|
|
synchronize_rcu();
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
if ((ACCESS_ONCE(sync_rcu_preempt_exp_count) - snap) > 0)
|
|
|
|
goto mb_ret; /* Others did our work for us. */
|
|
|
|
}
|
|
|
|
if ((ACCESS_ONCE(sync_rcu_preempt_exp_count) - snap) > 0)
|
|
|
|
goto unlock_mb_ret; /* Others did our work for us. */
|
|
|
|
|
2010-11-30 06:56:39 +01:00
|
|
|
/* force all RCU readers onto ->blkd_tasks lists. */
|
2009-12-02 21:10:15 +01:00
|
|
|
synchronize_sched_expedited();
|
|
|
|
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_lock_irqsave(&rsp->onofflock, flags);
|
2009-12-02 21:10:15 +01:00
|
|
|
|
|
|
|
/* Initialize ->expmask for all non-leaf rcu_node structures. */
|
|
|
|
rcu_for_each_nonleaf_node_breadth_first(rsp, rnp) {
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_lock(&rnp->lock); /* irqs already disabled. */
|
2009-12-02 21:10:15 +01:00
|
|
|
rnp->expmask = rnp->qsmaskinit;
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */
|
2009-12-02 21:10:15 +01:00
|
|
|
}
|
|
|
|
|
2010-11-30 06:56:39 +01:00
|
|
|
/* Snapshot current state of ->blkd_tasks lists. */
|
2009-12-02 21:10:15 +01:00
|
|
|
rcu_for_each_leaf_node(rsp, rnp)
|
|
|
|
sync_rcu_preempt_exp_init(rsp, rnp);
|
|
|
|
if (NUM_RCU_NODES > 1)
|
|
|
|
sync_rcu_preempt_exp_init(rsp, rcu_get_root(rsp));
|
|
|
|
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_unlock_irqrestore(&rsp->onofflock, flags);
|
2009-12-02 21:10:15 +01:00
|
|
|
|
2010-11-30 06:56:39 +01:00
|
|
|
/* Wait for snapshotted ->blkd_tasks lists to drain. */
|
2009-12-02 21:10:15 +01:00
|
|
|
rnp = rcu_get_root(rsp);
|
|
|
|
wait_event(sync_rcu_preempt_exp_wq,
|
|
|
|
sync_rcu_preempt_exp_done(rnp));
|
|
|
|
|
|
|
|
/* Clean up and exit. */
|
|
|
|
smp_mb(); /* ensure expedited GP seen before counter increment. */
|
|
|
|
ACCESS_ONCE(sync_rcu_preempt_exp_count)++;
|
|
|
|
unlock_mb_ret:
|
|
|
|
mutex_unlock(&sync_rcu_preempt_exp_mutex);
|
|
|
|
mb_ret:
|
|
|
|
smp_mb(); /* ensure subsequent action seen after grace period. */
|
2009-10-14 19:15:56 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(synchronize_rcu_expedited);
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Check to see if there is any immediate preemptible-RCU-related work
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
* to be done.
|
|
|
|
*/
|
|
|
|
static int rcu_preempt_pending(int cpu)
|
|
|
|
{
|
|
|
|
return __rcu_pending(&rcu_preempt_state,
|
|
|
|
&per_cpu(rcu_preempt_data, cpu));
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Does preemptible RCU need the CPU to stay out of dynticks mode?
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
*/
|
|
|
|
static int rcu_preempt_needs_cpu(int cpu)
|
|
|
|
{
|
|
|
|
return !!per_cpu(rcu_preempt_data, cpu).nxtlist;
|
|
|
|
}
|
|
|
|
|
2009-10-07 06:48:17 +02:00
|
|
|
/**
|
|
|
|
* rcu_barrier - Wait until all in-flight call_rcu() callbacks complete.
|
|
|
|
*/
|
|
|
|
void rcu_barrier(void)
|
|
|
|
{
|
|
|
|
_rcu_barrier(&rcu_preempt_state, call_rcu);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(rcu_barrier);
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Initialize preemptible RCU's per-CPU data.
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
*/
|
|
|
|
static void __cpuinit rcu_preempt_init_percpu_data(int cpu)
|
|
|
|
{
|
|
|
|
rcu_init_percpu_data(cpu, &rcu_preempt_state, 1);
|
|
|
|
}
|
|
|
|
|
2009-10-07 06:48:17 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Move preemptible RCU's callbacks from dying CPU to other online CPU.
|
2009-10-07 06:48:17 +02:00
|
|
|
*/
|
2010-10-20 08:13:06 +02:00
|
|
|
static void rcu_preempt_send_cbs_to_online(void)
|
2009-10-07 06:48:17 +02:00
|
|
|
{
|
2010-10-20 08:13:06 +02:00
|
|
|
rcu_send_cbs_to_online(&rcu_preempt_state);
|
2009-10-07 06:48:17 +02:00
|
|
|
}
|
|
|
|
|
2009-09-23 18:50:42 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Initialize preemptible RCU's state structures.
|
2009-09-23 18:50:42 +02:00
|
|
|
*/
|
|
|
|
static void __init __rcu_init_preempt(void)
|
|
|
|
{
|
2010-06-28 10:25:04 +02:00
|
|
|
rcu_init_one(&rcu_preempt_state, &rcu_preempt_data);
|
2009-09-23 18:50:42 +02:00
|
|
|
}
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Check for a task exiting while in a preemptible-RCU read-side
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
* critical section, clean up if so. No need to issue warnings,
|
|
|
|
* as debug_check_no_locks_held() already does this if lockdep
|
|
|
|
* is enabled.
|
|
|
|
*/
|
|
|
|
void exit_rcu(void)
|
|
|
|
{
|
|
|
|
struct task_struct *t = current;
|
|
|
|
|
|
|
|
if (t->rcu_read_lock_nesting == 0)
|
|
|
|
return;
|
|
|
|
t->rcu_read_lock_nesting = 1;
|
2011-02-25 20:37:59 +01:00
|
|
|
__rcu_read_unlock();
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
#else /* #ifdef CONFIG_TREE_PREEMPT_RCU */
|
|
|
|
|
2011-02-07 21:47:15 +01:00
|
|
|
static struct rcu_state *rcu_state = &rcu_sched_state;
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
/*
|
|
|
|
* Tell them what RCU they are running.
|
|
|
|
*/
|
2009-11-11 20:28:06 +01:00
|
|
|
static void __init rcu_bootup_announce(void)
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
{
|
|
|
|
printk(KERN_INFO "Hierarchical RCU implementation.\n");
|
2010-04-13 23:19:23 +02:00
|
|
|
rcu_bootup_announce_oddness();
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Return the number of RCU batches processed thus far for debug & stats.
|
|
|
|
*/
|
|
|
|
long rcu_batches_completed(void)
|
|
|
|
{
|
|
|
|
return rcu_batches_completed_sched();
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(rcu_batches_completed);
|
|
|
|
|
2010-01-05 00:09:10 +01:00
|
|
|
/*
|
|
|
|
* Force a quiescent state for RCU, which, because there is no preemptible
|
|
|
|
* RCU, becomes the same as rcu-sched.
|
|
|
|
*/
|
|
|
|
void rcu_force_quiescent_state(void)
|
|
|
|
{
|
|
|
|
rcu_sched_force_quiescent_state();
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(rcu_force_quiescent_state);
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because preemptible RCU does not exist, we never have to check for
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
* CPUs being in quiescent states.
|
|
|
|
*/
|
2009-09-13 18:15:10 +02:00
|
|
|
static void rcu_preempt_note_context_switch(int cpu)
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2009-09-23 18:50:41 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because preemptible RCU does not exist, there are never any preempted
|
2009-09-23 18:50:41 +02:00
|
|
|
* RCU readers.
|
|
|
|
*/
|
2011-02-07 21:47:15 +01:00
|
|
|
static int rcu_preempt_blocked_readers_cgp(struct rcu_node *rnp)
|
2009-09-23 18:50:41 +02:00
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
rcu: Fix grace-period-stall bug on large systems with CPU hotplug
When the last CPU of a given leaf rcu_node structure goes
offline, all of the tasks queued on that leaf rcu_node structure
(due to having blocked in their current RCU read-side critical
sections) are requeued onto the root rcu_node structure. This
requeuing is carried out by rcu_preempt_offline_tasks().
However, it is possible that these queued tasks are the only
thing preventing the leaf rcu_node structure from reporting a
quiescent state up the rcu_node hierarchy. Unfortunately, the
old code would fail to do this reporting, resulting in a
grace-period stall given the following sequence of events:
1. Kernel built for more than 32 CPUs on 32-bit systems or for more
than 64 CPUs on 64-bit systems, so that there is more than one
rcu_node structure. (Or CONFIG_RCU_FANOUT is artificially set
to a number smaller than CONFIG_NR_CPUS.)
2. The kernel is built with CONFIG_TREE_PREEMPT_RCU.
3. A task running on a CPU associated with a given leaf rcu_node
structure blocks while in an RCU read-side critical section
-and- that CPU has not yet passed through a quiescent state
for the current RCU grace period. This will cause the task
to be queued on the leaf rcu_node's blocked_tasks[] array, in
particular, on the element of this array corresponding to the
current grace period.
4. Each of the remaining CPUs corresponding to this same leaf rcu_node
structure pass through a quiescent state. However, the task is
still in its RCU read-side critical section, so these quiescent
states cannot be reported further up the rcu_node hierarchy.
Nevertheless, all bits in the leaf rcu_node structure's ->qsmask
field are now zero.
5. Each of the remaining CPUs go offline. (The events in step
#4 and #5 can happen in any order as long as each CPU passes
through a quiescent state before going offline.)
6. When the last CPU goes offline, __rcu_offline_cpu() will invoke
rcu_preempt_offline_tasks(), which will move the task to the
root rcu_node structure, but without reporting a quiescent state
up the rcu_node hierarchy (and this failure to report a quiescent
state is the bug).
But because this leaf rcu_node structure's ->qsmask field is
already zero and its ->block_tasks[] entries are all empty,
force_quiescent_state() will skip this rcu_node structure.
Therefore, grace periods are now hung.
This patch abstracts some code out of rcu_read_unlock_special(),
calling the result task_quiet() by analogy with cpu_quiet(), and
invokes task_quiet() from both rcu_read_lock_special() and
__rcu_offline_cpu(). Invoking task_quiet() from
__rcu_offline_cpu() reports the quiescent state up the rcu_node
hierarchy, fixing the bug. This ends up requiring a separate
lock_class_key per level of the rcu_node hierarchy, which this
patch also provides.
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: <12589088301770-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-22 17:53:48 +01:00
|
|
|
#ifdef CONFIG_HOTPLUG_CPU
|
|
|
|
|
|
|
|
/* Because preemptible RCU does not exist, no quieting of tasks. */
|
2009-12-02 21:10:13 +01:00
|
|
|
static void rcu_report_unblock_qs_rnp(struct rcu_node *rnp, unsigned long flags)
|
rcu: Fix grace-period-stall bug on large systems with CPU hotplug
When the last CPU of a given leaf rcu_node structure goes
offline, all of the tasks queued on that leaf rcu_node structure
(due to having blocked in their current RCU read-side critical
sections) are requeued onto the root rcu_node structure. This
requeuing is carried out by rcu_preempt_offline_tasks().
However, it is possible that these queued tasks are the only
thing preventing the leaf rcu_node structure from reporting a
quiescent state up the rcu_node hierarchy. Unfortunately, the
old code would fail to do this reporting, resulting in a
grace-period stall given the following sequence of events:
1. Kernel built for more than 32 CPUs on 32-bit systems or for more
than 64 CPUs on 64-bit systems, so that there is more than one
rcu_node structure. (Or CONFIG_RCU_FANOUT is artificially set
to a number smaller than CONFIG_NR_CPUS.)
2. The kernel is built with CONFIG_TREE_PREEMPT_RCU.
3. A task running on a CPU associated with a given leaf rcu_node
structure blocks while in an RCU read-side critical section
-and- that CPU has not yet passed through a quiescent state
for the current RCU grace period. This will cause the task
to be queued on the leaf rcu_node's blocked_tasks[] array, in
particular, on the element of this array corresponding to the
current grace period.
4. Each of the remaining CPUs corresponding to this same leaf rcu_node
structure pass through a quiescent state. However, the task is
still in its RCU read-side critical section, so these quiescent
states cannot be reported further up the rcu_node hierarchy.
Nevertheless, all bits in the leaf rcu_node structure's ->qsmask
field are now zero.
5. Each of the remaining CPUs go offline. (The events in step
#4 and #5 can happen in any order as long as each CPU passes
through a quiescent state before going offline.)
6. When the last CPU goes offline, __rcu_offline_cpu() will invoke
rcu_preempt_offline_tasks(), which will move the task to the
root rcu_node structure, but without reporting a quiescent state
up the rcu_node hierarchy (and this failure to report a quiescent
state is the bug).
But because this leaf rcu_node structure's ->qsmask field is
already zero and its ->block_tasks[] entries are all empty,
force_quiescent_state() will skip this rcu_node structure.
Therefore, grace periods are now hung.
This patch abstracts some code out of rcu_read_unlock_special(),
calling the result task_quiet() by analogy with cpu_quiet(), and
invokes task_quiet() from both rcu_read_lock_special() and
__rcu_offline_cpu(). Invoking task_quiet() from
__rcu_offline_cpu() reports the quiescent state up the rcu_node
hierarchy, fixing the bug. This ends up requiring a separate
lock_class_key per level of the rcu_node hierarchy, which this
patch also provides.
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: <12589088301770-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-22 17:53:48 +01:00
|
|
|
{
|
2010-02-23 02:05:02 +01:00
|
|
|
raw_spin_unlock_irqrestore(&rnp->lock, flags);
|
rcu: Fix grace-period-stall bug on large systems with CPU hotplug
When the last CPU of a given leaf rcu_node structure goes
offline, all of the tasks queued on that leaf rcu_node structure
(due to having blocked in their current RCU read-side critical
sections) are requeued onto the root rcu_node structure. This
requeuing is carried out by rcu_preempt_offline_tasks().
However, it is possible that these queued tasks are the only
thing preventing the leaf rcu_node structure from reporting a
quiescent state up the rcu_node hierarchy. Unfortunately, the
old code would fail to do this reporting, resulting in a
grace-period stall given the following sequence of events:
1. Kernel built for more than 32 CPUs on 32-bit systems or for more
than 64 CPUs on 64-bit systems, so that there is more than one
rcu_node structure. (Or CONFIG_RCU_FANOUT is artificially set
to a number smaller than CONFIG_NR_CPUS.)
2. The kernel is built with CONFIG_TREE_PREEMPT_RCU.
3. A task running on a CPU associated with a given leaf rcu_node
structure blocks while in an RCU read-side critical section
-and- that CPU has not yet passed through a quiescent state
for the current RCU grace period. This will cause the task
to be queued on the leaf rcu_node's blocked_tasks[] array, in
particular, on the element of this array corresponding to the
current grace period.
4. Each of the remaining CPUs corresponding to this same leaf rcu_node
structure pass through a quiescent state. However, the task is
still in its RCU read-side critical section, so these quiescent
states cannot be reported further up the rcu_node hierarchy.
Nevertheless, all bits in the leaf rcu_node structure's ->qsmask
field are now zero.
5. Each of the remaining CPUs go offline. (The events in step
#4 and #5 can happen in any order as long as each CPU passes
through a quiescent state before going offline.)
6. When the last CPU goes offline, __rcu_offline_cpu() will invoke
rcu_preempt_offline_tasks(), which will move the task to the
root rcu_node structure, but without reporting a quiescent state
up the rcu_node hierarchy (and this failure to report a quiescent
state is the bug).
But because this leaf rcu_node structure's ->qsmask field is
already zero and its ->block_tasks[] entries are all empty,
force_quiescent_state() will skip this rcu_node structure.
Therefore, grace periods are now hung.
This patch abstracts some code out of rcu_read_unlock_special(),
calling the result task_quiet() by analogy with cpu_quiet(), and
invokes task_quiet() from both rcu_read_lock_special() and
__rcu_offline_cpu(). Invoking task_quiet() from
__rcu_offline_cpu() reports the quiescent state up the rcu_node
hierarchy, fixing the bug. This ends up requiring a separate
lock_class_key per level of the rcu_node hierarchy, which this
patch also provides.
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: <12589088301770-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-22 17:53:48 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
#endif /* #ifdef CONFIG_HOTPLUG_CPU */
|
|
|
|
|
2010-02-23 02:05:05 +01:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because preemptible RCU does not exist, we never have to check for
|
2010-02-23 02:05:05 +01:00
|
|
|
* tasks blocked within RCU read-side critical sections.
|
|
|
|
*/
|
|
|
|
static void rcu_print_detail_task_stall(struct rcu_state *rsp)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because preemptible RCU does not exist, we never have to check for
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
* tasks blocked within RCU read-side critical sections.
|
|
|
|
*/
|
|
|
|
static void rcu_print_task_stall(struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2010-08-10 23:28:53 +02:00
|
|
|
/*
|
|
|
|
* Because preemptible RCU does not exist, there is no need to suppress
|
|
|
|
* its CPU stall warnings.
|
|
|
|
*/
|
|
|
|
static void rcu_preempt_stall_reset(void)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2009-09-13 18:15:09 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because there is no preemptible RCU, there can be no readers blocked,
|
2009-09-18 18:50:19 +02:00
|
|
|
* so there is no need to check for blocked tasks. So check only for
|
|
|
|
* bogus qsmask values.
|
2009-09-13 18:15:09 +02:00
|
|
|
*/
|
|
|
|
static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp)
|
|
|
|
{
|
2009-09-18 18:50:19 +02:00
|
|
|
WARN_ON_ONCE(rnp->qsmask);
|
2009-09-13 18:15:09 +02:00
|
|
|
}
|
|
|
|
|
2009-08-24 18:42:01 +02:00
|
|
|
#ifdef CONFIG_HOTPLUG_CPU
|
|
|
|
|
2009-08-27 23:58:16 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because preemptible RCU does not exist, it never needs to migrate
|
rcu: Fix TREE_PREEMPT_RCU CPU_HOTPLUG bad-luck hang
If the following sequence of events occurs, then
TREE_PREEMPT_RCU will hang waiting for a grace period to
complete, eventually OOMing the system:
o A TREE_PREEMPT_RCU build of the kernel is booted on a system
with more than 64 physical CPUs present (32 on a 32-bit system).
Alternatively, a TREE_PREEMPT_RCU build of the kernel is booted
with RCU_FANOUT set to a sufficiently small value that the
physical CPUs populate two or more leaf rcu_node structures.
o A task is preempted in an RCU read-side critical section
while running on a CPU corresponding to a given leaf rcu_node
structure.
o All CPUs corresponding to this same leaf rcu_node structure
record quiescent states for the current grace period.
o All of these same CPUs go offline (hence the need for enough
physical CPUs to populate more than one leaf rcu_node structure).
This causes the preempted task to be moved to the root rcu_node
structure.
At this point, there is nothing left to cause the quiescent
state to be propagated up the rcu_node tree, so the current
grace period never completes.
The simplest fix, especially after considering the deadlock
possibilities, is to detect this situation when the last CPU is
offlined, and to set that CPU's ->qsmask bit in its leaf
rcu_node structure. This will cause the next invocation of
force_quiescent_state() to end the grace period.
Without this fix, this hang can be triggered in an hour or so on
some machines with rcutorture and random CPU onlining/offlining.
With this fix, these same machines pass a full 10 hours of this
sort of abuse.
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: <20091015162614.GA19131@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-15 18:26:14 +02:00
|
|
|
* tasks that were blocked within RCU read-side critical sections, and
|
|
|
|
* such non-existent tasks cannot possibly have been blocking the current
|
|
|
|
* grace period.
|
2009-08-27 23:58:16 +02:00
|
|
|
*/
|
rcu: Fix TREE_PREEMPT_RCU CPU_HOTPLUG bad-luck hang
If the following sequence of events occurs, then
TREE_PREEMPT_RCU will hang waiting for a grace period to
complete, eventually OOMing the system:
o A TREE_PREEMPT_RCU build of the kernel is booted on a system
with more than 64 physical CPUs present (32 on a 32-bit system).
Alternatively, a TREE_PREEMPT_RCU build of the kernel is booted
with RCU_FANOUT set to a sufficiently small value that the
physical CPUs populate two or more leaf rcu_node structures.
o A task is preempted in an RCU read-side critical section
while running on a CPU corresponding to a given leaf rcu_node
structure.
o All CPUs corresponding to this same leaf rcu_node structure
record quiescent states for the current grace period.
o All of these same CPUs go offline (hence the need for enough
physical CPUs to populate more than one leaf rcu_node structure).
This causes the preempted task to be moved to the root rcu_node
structure.
At this point, there is nothing left to cause the quiescent
state to be propagated up the rcu_node tree, so the current
grace period never completes.
The simplest fix, especially after considering the deadlock
possibilities, is to detect this situation when the last CPU is
offlined, and to set that CPU's ->qsmask bit in its leaf
rcu_node structure. This will cause the next invocation of
force_quiescent_state() to end the grace period.
Without this fix, this hang can be triggered in an hour or so on
some machines with rcutorture and random CPU onlining/offlining.
With this fix, these same machines pass a full 10 hours of this
sort of abuse.
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: <20091015162614.GA19131@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-15 18:26:14 +02:00
|
|
|
static int rcu_preempt_offline_tasks(struct rcu_state *rsp,
|
|
|
|
struct rcu_node *rnp,
|
|
|
|
struct rcu_data *rdp)
|
2009-08-27 23:58:16 +02:00
|
|
|
{
|
rcu: Fix TREE_PREEMPT_RCU CPU_HOTPLUG bad-luck hang
If the following sequence of events occurs, then
TREE_PREEMPT_RCU will hang waiting for a grace period to
complete, eventually OOMing the system:
o A TREE_PREEMPT_RCU build of the kernel is booted on a system
with more than 64 physical CPUs present (32 on a 32-bit system).
Alternatively, a TREE_PREEMPT_RCU build of the kernel is booted
with RCU_FANOUT set to a sufficiently small value that the
physical CPUs populate two or more leaf rcu_node structures.
o A task is preempted in an RCU read-side critical section
while running on a CPU corresponding to a given leaf rcu_node
structure.
o All CPUs corresponding to this same leaf rcu_node structure
record quiescent states for the current grace period.
o All of these same CPUs go offline (hence the need for enough
physical CPUs to populate more than one leaf rcu_node structure).
This causes the preempted task to be moved to the root rcu_node
structure.
At this point, there is nothing left to cause the quiescent
state to be propagated up the rcu_node tree, so the current
grace period never completes.
The simplest fix, especially after considering the deadlock
possibilities, is to detect this situation when the last CPU is
offlined, and to set that CPU's ->qsmask bit in its leaf
rcu_node structure. This will cause the next invocation of
force_quiescent_state() to end the grace period.
Without this fix, this hang can be triggered in an hour or so on
some machines with rcutorture and random CPU onlining/offlining.
With this fix, these same machines pass a full 10 hours of this
sort of abuse.
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: <20091015162614.GA19131@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-15 18:26:14 +02:00
|
|
|
return 0;
|
2009-08-27 23:58:16 +02:00
|
|
|
}
|
|
|
|
|
2009-08-24 18:42:01 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because preemptible RCU does not exist, it never needs CPU-offline
|
2009-08-24 18:42:01 +02:00
|
|
|
* processing.
|
|
|
|
*/
|
|
|
|
static void rcu_preempt_offline_cpu(int cpu)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
#endif /* #ifdef CONFIG_HOTPLUG_CPU */
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because preemptible RCU does not exist, it never has any callbacks
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
* to check.
|
|
|
|
*/
|
2009-09-23 18:50:42 +02:00
|
|
|
static void rcu_preempt_check_callbacks(int cpu)
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because preemptible RCU does not exist, it never has any callbacks
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
* to process.
|
|
|
|
*/
|
2009-09-23 18:50:42 +02:00
|
|
|
static void rcu_preempt_process_callbacks(void)
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2009-10-14 19:15:56 +02:00
|
|
|
/*
|
|
|
|
* Wait for an rcu-preempt grace period, but make it happen quickly.
|
2011-03-02 22:15:15 +01:00
|
|
|
* But because preemptible RCU does not exist, map to rcu-sched.
|
2009-10-14 19:15:56 +02:00
|
|
|
*/
|
|
|
|
void synchronize_rcu_expedited(void)
|
|
|
|
{
|
|
|
|
synchronize_sched_expedited();
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(synchronize_rcu_expedited);
|
|
|
|
|
2009-12-02 21:10:15 +01:00
|
|
|
#ifdef CONFIG_HOTPLUG_CPU
|
|
|
|
|
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because preemptible RCU does not exist, there is never any need to
|
2009-12-02 21:10:15 +01:00
|
|
|
* report on tasks preempted in RCU read-side critical sections during
|
|
|
|
* expedited RCU grace periods.
|
|
|
|
*/
|
|
|
|
static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
#endif /* #ifdef CONFIG_HOTPLUG_CPU */
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because preemptible RCU does not exist, it never has any work to do.
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
*/
|
|
|
|
static int rcu_preempt_pending(int cpu)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because preemptible RCU does not exist, it never needs any CPU.
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
*/
|
|
|
|
static int rcu_preempt_needs_cpu(int cpu)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-10-07 06:48:17 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because preemptible RCU does not exist, rcu_barrier() is just
|
2009-10-07 06:48:17 +02:00
|
|
|
* another name for rcu_barrier_sched().
|
|
|
|
*/
|
|
|
|
void rcu_barrier(void)
|
|
|
|
{
|
|
|
|
rcu_barrier_sched();
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(rcu_barrier);
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because preemptible RCU does not exist, there is no per-CPU
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
* data to initialize.
|
|
|
|
*/
|
|
|
|
static void __cpuinit rcu_preempt_init_percpu_data(int cpu)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2009-10-07 06:48:17 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because there is no preemptible RCU, there are no callbacks to move.
|
2009-10-07 06:48:17 +02:00
|
|
|
*/
|
2010-10-20 08:13:06 +02:00
|
|
|
static void rcu_preempt_send_cbs_to_online(void)
|
2009-10-07 06:48:17 +02:00
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2009-09-23 18:50:42 +02:00
|
|
|
/*
|
2011-03-02 22:15:15 +01:00
|
|
|
* Because preemptible RCU does not exist, it need not be initialized.
|
2009-09-23 18:50:42 +02:00
|
|
|
*/
|
|
|
|
static void __init __rcu_init_preempt(void)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 22:56:52 +02:00
|
|
|
#endif /* #else #ifdef CONFIG_TREE_PREEMPT_RCU */
|
2010-02-23 02:04:59 +01:00
|
|
|
|
2011-02-07 21:47:15 +01:00
|
|
|
#ifdef CONFIG_RCU_BOOST
|
|
|
|
|
|
|
|
#include "rtmutex_common.h"
|
|
|
|
|
2011-02-22 22:42:43 +01:00
|
|
|
#ifdef CONFIG_RCU_TRACE
|
|
|
|
|
|
|
|
static void rcu_initiate_boost_trace(struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
if (list_empty(&rnp->blkd_tasks))
|
|
|
|
rnp->n_balk_blkd_tasks++;
|
|
|
|
else if (rnp->exp_tasks == NULL && rnp->gp_tasks == NULL)
|
|
|
|
rnp->n_balk_exp_gp_tasks++;
|
|
|
|
else if (rnp->gp_tasks != NULL && rnp->boost_tasks != NULL)
|
|
|
|
rnp->n_balk_boost_tasks++;
|
|
|
|
else if (rnp->gp_tasks != NULL && rnp->qsmask != 0)
|
|
|
|
rnp->n_balk_notblocked++;
|
|
|
|
else if (rnp->gp_tasks != NULL &&
|
2011-05-02 12:46:10 +02:00
|
|
|
ULONG_CMP_LT(jiffies, rnp->boost_time))
|
2011-02-22 22:42:43 +01:00
|
|
|
rnp->n_balk_notyet++;
|
|
|
|
else
|
|
|
|
rnp->n_balk_nos++;
|
|
|
|
}
|
|
|
|
|
|
|
|
#else /* #ifdef CONFIG_RCU_TRACE */
|
|
|
|
|
|
|
|
static void rcu_initiate_boost_trace(struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
#endif /* #else #ifdef CONFIG_RCU_TRACE */
|
|
|
|
|
2011-02-07 21:47:15 +01:00
|
|
|
/*
|
|
|
|
* Carry out RCU priority boosting on the task indicated by ->exp_tasks
|
|
|
|
* or ->boost_tasks, advancing the pointer to the next task in the
|
|
|
|
* ->blkd_tasks list.
|
|
|
|
*
|
|
|
|
* Note that irqs must be enabled: boosting the task can block.
|
|
|
|
* Returns 1 if there are more tasks needing to be boosted.
|
|
|
|
*/
|
|
|
|
static int rcu_boost(struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
struct rt_mutex mtx;
|
|
|
|
struct task_struct *t;
|
|
|
|
struct list_head *tb;
|
|
|
|
|
|
|
|
if (rnp->exp_tasks == NULL && rnp->boost_tasks == NULL)
|
|
|
|
return 0; /* Nothing left to boost. */
|
|
|
|
|
|
|
|
raw_spin_lock_irqsave(&rnp->lock, flags);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Recheck under the lock: all tasks in need of boosting
|
|
|
|
* might exit their RCU read-side critical sections on their own.
|
|
|
|
*/
|
|
|
|
if (rnp->exp_tasks == NULL && rnp->boost_tasks == NULL) {
|
|
|
|
raw_spin_unlock_irqrestore(&rnp->lock, flags);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Preferentially boost tasks blocking expedited grace periods.
|
|
|
|
* This cannot starve the normal grace periods because a second
|
|
|
|
* expedited grace period must boost all blocked tasks, including
|
|
|
|
* those blocking the pre-existing normal grace period.
|
|
|
|
*/
|
2011-02-22 22:42:43 +01:00
|
|
|
if (rnp->exp_tasks != NULL) {
|
2011-02-07 21:47:15 +01:00
|
|
|
tb = rnp->exp_tasks;
|
2011-02-22 22:42:43 +01:00
|
|
|
rnp->n_exp_boosts++;
|
|
|
|
} else {
|
2011-02-07 21:47:15 +01:00
|
|
|
tb = rnp->boost_tasks;
|
2011-02-22 22:42:43 +01:00
|
|
|
rnp->n_normal_boosts++;
|
|
|
|
}
|
|
|
|
rnp->n_tasks_boosted++;
|
2011-02-07 21:47:15 +01:00
|
|
|
|
|
|
|
/*
|
|
|
|
* We boost task t by manufacturing an rt_mutex that appears to
|
|
|
|
* be held by task t. We leave a pointer to that rt_mutex where
|
|
|
|
* task t can find it, and task t will release the mutex when it
|
|
|
|
* exits its outermost RCU read-side critical section. Then
|
|
|
|
* simply acquiring this artificial rt_mutex will boost task
|
|
|
|
* t's priority. (Thanks to tglx for suggesting this approach!)
|
|
|
|
*
|
|
|
|
* Note that task t must acquire rnp->lock to remove itself from
|
|
|
|
* the ->blkd_tasks list, which it will do from exit() if from
|
|
|
|
* nowhere else. We therefore are guaranteed that task t will
|
|
|
|
* stay around at least until we drop rnp->lock. Note that
|
|
|
|
* rnp->lock also resolves races between our priority boosting
|
|
|
|
* and task t's exiting its outermost RCU read-side critical
|
|
|
|
* section.
|
|
|
|
*/
|
|
|
|
t = container_of(tb, struct task_struct, rcu_node_entry);
|
|
|
|
rt_mutex_init_proxy_locked(&mtx, t);
|
|
|
|
t->rcu_boost_mutex = &mtx;
|
|
|
|
t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BOOSTED;
|
|
|
|
raw_spin_unlock_irqrestore(&rnp->lock, flags);
|
|
|
|
rt_mutex_lock(&mtx); /* Side effect: boosts task t's priority. */
|
|
|
|
rt_mutex_unlock(&mtx); /* Keep lockdep happy. */
|
|
|
|
|
|
|
|
return rnp->exp_tasks != NULL || rnp->boost_tasks != NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Timer handler to initiate waking up of boost kthreads that
|
|
|
|
* have yielded the CPU due to excessive numbers of tasks to
|
|
|
|
* boost. We wake up the per-rcu_node kthread, which in turn
|
|
|
|
* will wake up the booster kthread.
|
|
|
|
*/
|
|
|
|
static void rcu_boost_kthread_timer(unsigned long arg)
|
|
|
|
{
|
2011-05-05 06:43:49 +02:00
|
|
|
invoke_rcu_node_kthread((struct rcu_node *)arg);
|
2011-02-07 21:47:15 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Priority-boosting kthread. One per leaf rcu_node and one for the
|
|
|
|
* root rcu_node.
|
|
|
|
*/
|
|
|
|
static int rcu_boost_kthread(void *arg)
|
|
|
|
{
|
|
|
|
struct rcu_node *rnp = (struct rcu_node *)arg;
|
|
|
|
int spincnt = 0;
|
|
|
|
int more2boost;
|
|
|
|
|
|
|
|
for (;;) {
|
2011-03-30 02:48:28 +02:00
|
|
|
rnp->boost_kthread_status = RCU_KTHREAD_WAITING;
|
2011-05-21 01:06:29 +02:00
|
|
|
rcu_wait(rnp->boost_tasks || rnp->exp_tasks);
|
2011-03-30 02:48:28 +02:00
|
|
|
rnp->boost_kthread_status = RCU_KTHREAD_RUNNING;
|
2011-02-07 21:47:15 +01:00
|
|
|
more2boost = rcu_boost(rnp);
|
|
|
|
if (more2boost)
|
|
|
|
spincnt++;
|
|
|
|
else
|
|
|
|
spincnt = 0;
|
|
|
|
if (spincnt > 10) {
|
|
|
|
rcu_yield(rcu_boost_kthread_timer, (unsigned long)rnp);
|
|
|
|
spincnt = 0;
|
|
|
|
}
|
|
|
|
}
|
2011-05-05 06:43:49 +02:00
|
|
|
/* NOTREACHED */
|
2011-02-07 21:47:15 +01:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Check to see if it is time to start boosting RCU readers that are
|
|
|
|
* blocking the current grace period, and, if so, tell the per-rcu_node
|
|
|
|
* kthread to start boosting them. If there is an expedited grace
|
|
|
|
* period in progress, it is always time to boost.
|
|
|
|
*
|
2011-05-05 06:43:49 +02:00
|
|
|
* The caller must hold rnp->lock, which this function releases,
|
|
|
|
* but irqs remain disabled. The ->boost_kthread_task is immortal,
|
|
|
|
* so we don't need to worry about it going away.
|
2011-02-07 21:47:15 +01:00
|
|
|
*/
|
2011-05-05 06:43:49 +02:00
|
|
|
static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags)
|
2011-02-07 21:47:15 +01:00
|
|
|
{
|
|
|
|
struct task_struct *t;
|
|
|
|
|
2011-02-22 22:42:43 +01:00
|
|
|
if (!rcu_preempt_blocked_readers_cgp(rnp) && rnp->exp_tasks == NULL) {
|
|
|
|
rnp->n_balk_exp_gp_tasks++;
|
2011-05-05 06:43:49 +02:00
|
|
|
raw_spin_unlock_irqrestore(&rnp->lock, flags);
|
2011-02-07 21:47:15 +01:00
|
|
|
return;
|
2011-02-22 22:42:43 +01:00
|
|
|
}
|
2011-02-07 21:47:15 +01:00
|
|
|
if (rnp->exp_tasks != NULL ||
|
|
|
|
(rnp->gp_tasks != NULL &&
|
|
|
|
rnp->boost_tasks == NULL &&
|
|
|
|
rnp->qsmask == 0 &&
|
|
|
|
ULONG_CMP_GE(jiffies, rnp->boost_time))) {
|
|
|
|
if (rnp->exp_tasks == NULL)
|
|
|
|
rnp->boost_tasks = rnp->gp_tasks;
|
2011-05-05 06:43:49 +02:00
|
|
|
raw_spin_unlock_irqrestore(&rnp->lock, flags);
|
2011-02-07 21:47:15 +01:00
|
|
|
t = rnp->boost_kthread_task;
|
|
|
|
if (t != NULL)
|
|
|
|
wake_up_process(t);
|
2011-05-05 06:43:49 +02:00
|
|
|
} else {
|
2011-02-22 22:42:43 +01:00
|
|
|
rcu_initiate_boost_trace(rnp);
|
2011-05-05 06:43:49 +02:00
|
|
|
raw_spin_unlock_irqrestore(&rnp->lock, flags);
|
|
|
|
}
|
2011-02-07 21:47:15 +01:00
|
|
|
}
|
|
|
|
|
2011-04-14 21:13:53 +02:00
|
|
|
/*
|
|
|
|
* Set the affinity of the boost kthread. The CPU-hotplug locks are
|
|
|
|
* held, so no one should be messing with the existence of the boost
|
|
|
|
* kthread.
|
|
|
|
*/
|
2011-02-07 21:47:15 +01:00
|
|
|
static void rcu_boost_kthread_setaffinity(struct rcu_node *rnp,
|
|
|
|
cpumask_var_t cm)
|
|
|
|
{
|
|
|
|
struct task_struct *t;
|
|
|
|
|
|
|
|
t = rnp->boost_kthread_task;
|
|
|
|
if (t != NULL)
|
|
|
|
set_cpus_allowed_ptr(rnp->boost_kthread_task, cm);
|
|
|
|
}
|
|
|
|
|
|
|
|
#define RCU_BOOST_DELAY_JIFFIES DIV_ROUND_UP(CONFIG_RCU_BOOST_DELAY * HZ, 1000)
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Do priority-boost accounting for the start of a new grace period.
|
|
|
|
*/
|
|
|
|
static void rcu_preempt_boost_start_gp(struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
rnp->boost_time = jiffies + RCU_BOOST_DELAY_JIFFIES;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Create an RCU-boost kthread for the specified node if one does not
|
|
|
|
* already exist. We only create this kthread for preemptible RCU.
|
|
|
|
* Returns zero if all is well, a negated errno otherwise.
|
|
|
|
*/
|
|
|
|
static int __cpuinit rcu_spawn_one_boost_kthread(struct rcu_state *rsp,
|
|
|
|
struct rcu_node *rnp,
|
|
|
|
int rnp_index)
|
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
struct sched_param sp;
|
|
|
|
struct task_struct *t;
|
|
|
|
|
|
|
|
if (&rcu_preempt_state != rsp)
|
|
|
|
return 0;
|
|
|
|
if (rnp->boost_kthread_task != NULL)
|
|
|
|
return 0;
|
|
|
|
t = kthread_create(rcu_boost_kthread, (void *)rnp,
|
|
|
|
"rcub%d", rnp_index);
|
|
|
|
if (IS_ERR(t))
|
|
|
|
return PTR_ERR(t);
|
|
|
|
raw_spin_lock_irqsave(&rnp->lock, flags);
|
|
|
|
rnp->boost_kthread_task = t;
|
|
|
|
raw_spin_unlock_irqrestore(&rnp->lock, flags);
|
|
|
|
sp.sched_priority = RCU_KTHREAD_PRIO;
|
|
|
|
sched_setscheduler_nocheck(t, SCHED_FIFO, &sp);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2011-05-30 13:34:51 +02:00
|
|
|
static void __cpuinit rcu_wake_one_boost_kthread(struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
if (rnp->boost_kthread_task)
|
|
|
|
wake_up_process(rnp->boost_kthread_task);
|
|
|
|
}
|
|
|
|
|
2011-02-07 21:47:15 +01:00
|
|
|
#else /* #ifdef CONFIG_RCU_BOOST */
|
|
|
|
|
2011-05-05 06:43:49 +02:00
|
|
|
static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags)
|
2011-02-07 21:47:15 +01:00
|
|
|
{
|
2011-05-05 06:43:49 +02:00
|
|
|
raw_spin_unlock_irqrestore(&rnp->lock, flags);
|
2011-02-07 21:47:15 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
static void rcu_boost_kthread_setaffinity(struct rcu_node *rnp,
|
|
|
|
cpumask_var_t cm)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
static void rcu_preempt_boost_start_gp(struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
static int __cpuinit rcu_spawn_one_boost_kthread(struct rcu_state *rsp,
|
|
|
|
struct rcu_node *rnp,
|
|
|
|
int rnp_index)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2011-05-30 13:34:51 +02:00
|
|
|
static void __cpuinit rcu_wake_one_boost_kthread(struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2011-02-07 21:47:15 +01:00
|
|
|
#endif /* #else #ifdef CONFIG_RCU_BOOST */
|
|
|
|
|
2010-10-21 05:29:05 +02:00
|
|
|
#ifndef CONFIG_SMP
|
|
|
|
|
|
|
|
void synchronize_sched_expedited(void)
|
|
|
|
{
|
|
|
|
cond_resched();
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(synchronize_sched_expedited);
|
|
|
|
|
|
|
|
#else /* #ifndef CONFIG_SMP */
|
|
|
|
|
2010-11-23 06:36:11 +01:00
|
|
|
static atomic_t sync_sched_expedited_started = ATOMIC_INIT(0);
|
|
|
|
static atomic_t sync_sched_expedited_done = ATOMIC_INIT(0);
|
2010-10-21 05:29:05 +02:00
|
|
|
|
|
|
|
static int synchronize_sched_expedited_cpu_stop(void *data)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* There must be a full memory barrier on each affected CPU
|
|
|
|
* between the time that try_stop_cpus() is called and the
|
|
|
|
* time that it returns.
|
|
|
|
*
|
|
|
|
* In the current initial implementation of cpu_stop, the
|
|
|
|
* above condition is already met when the control reaches
|
|
|
|
* this point and the following smp_mb() is not strictly
|
|
|
|
* necessary. Do smp_mb() anyway for documentation and
|
|
|
|
* robustness against future implementation changes.
|
|
|
|
*/
|
|
|
|
smp_mb(); /* See above comment block. */
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Wait for an rcu-sched grace period to elapse, but use "big hammer"
|
|
|
|
* approach to force grace period to end quickly. This consumes
|
|
|
|
* significant time on all CPUs, and is thus not recommended for
|
|
|
|
* any sort of common-case code.
|
|
|
|
*
|
|
|
|
* Note that it is illegal to call this function while holding any
|
|
|
|
* lock that is acquired by a CPU-hotplug notifier. Failing to
|
|
|
|
* observe this restriction will result in deadlock.
|
rcu: fix race condition in synchronize_sched_expedited()
The new (early 2010) implementation of synchronize_sched_expedited() uses
try_stop_cpu() to force a context switch on every CPU. It also permits
concurrent calls to synchronize_sched_expedited() to share a single call
to try_stop_cpu() through use of an atomically incremented
synchronize_sched_expedited_count variable. Unfortunately, this is
subject to failure as follows:
o Task A invokes synchronize_sched_expedited(), try_stop_cpus()
succeeds, but Task A is preempted before getting to the atomic
increment of synchronize_sched_expedited_count.
o Task B also invokes synchronize_sched_expedited(), with exactly
the same outcome as Task A.
o Task C also invokes synchronize_sched_expedited(), again with
exactly the same outcome as Tasks A and B.
o Task D also invokes synchronize_sched_expedited(), but only
gets as far as acquiring the mutex within try_stop_cpus()
before being preempted, interrupted, or otherwise delayed.
o Task E also invokes synchronize_sched_expedited(), but only
gets to the snapshotting of synchronize_sched_expedited_count.
o Tasks A, B, and C all increment synchronize_sched_expedited_count.
o Task E fails to get the mutex, so checks the new value
of synchronize_sched_expedited_count. It finds that the
value has increased, so (wrongly) assumes that its work
has been done, returning despite there having been no
expedited grace period since it began.
The solution is to have the lowest-numbered CPU atomically increment
the synchronize_sched_expedited_count variable within the
synchronize_sched_expedited_cpu_stop() function, which is under
the protection of the mutex acquired by try_stop_cpus(). However, this
also requires that piggybacking tasks wait for three rather than two
instances of try_stop_cpu(), because we cannot control the order in
which the per-CPU callback function occur.
Cc: Tejun Heo <tj@kernel.org>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2010-10-25 16:39:22 +02:00
|
|
|
*
|
2010-11-23 06:36:11 +01:00
|
|
|
* This implementation can be thought of as an application of ticket
|
|
|
|
* locking to RCU, with sync_sched_expedited_started and
|
|
|
|
* sync_sched_expedited_done taking on the roles of the halves
|
|
|
|
* of the ticket-lock word. Each task atomically increments
|
|
|
|
* sync_sched_expedited_started upon entry, snapshotting the old value,
|
|
|
|
* then attempts to stop all the CPUs. If this succeeds, then each
|
|
|
|
* CPU will have executed a context switch, resulting in an RCU-sched
|
|
|
|
* grace period. We are then done, so we use atomic_cmpxchg() to
|
|
|
|
* update sync_sched_expedited_done to match our snapshot -- but
|
|
|
|
* only if someone else has not already advanced past our snapshot.
|
|
|
|
*
|
|
|
|
* On the other hand, if try_stop_cpus() fails, we check the value
|
|
|
|
* of sync_sched_expedited_done. If it has advanced past our
|
|
|
|
* initial snapshot, then someone else must have forced a grace period
|
|
|
|
* some time after we took our snapshot. In this case, our work is
|
|
|
|
* done for us, and we can simply return. Otherwise, we try again,
|
|
|
|
* but keep our initial snapshot for purposes of checking for someone
|
|
|
|
* doing our work for us.
|
|
|
|
*
|
|
|
|
* If we fail too many times in a row, we fall back to synchronize_sched().
|
2010-10-21 05:29:05 +02:00
|
|
|
*/
|
|
|
|
void synchronize_sched_expedited(void)
|
|
|
|
{
|
2010-11-23 06:36:11 +01:00
|
|
|
int firstsnap, s, snap, trycount = 0;
|
2010-10-21 05:29:05 +02:00
|
|
|
|
2010-11-23 06:36:11 +01:00
|
|
|
/* Note that atomic_inc_return() implies full memory barrier. */
|
|
|
|
firstsnap = snap = atomic_inc_return(&sync_sched_expedited_started);
|
2010-10-21 05:29:05 +02:00
|
|
|
get_online_cpus();
|
2010-11-23 06:36:11 +01:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Each pass through the following loop attempts to force a
|
|
|
|
* context switch on each CPU.
|
|
|
|
*/
|
2010-10-21 05:29:05 +02:00
|
|
|
while (try_stop_cpus(cpu_online_mask,
|
|
|
|
synchronize_sched_expedited_cpu_stop,
|
|
|
|
NULL) == -EAGAIN) {
|
|
|
|
put_online_cpus();
|
2010-11-23 06:36:11 +01:00
|
|
|
|
|
|
|
/* No joy, try again later. Or just synchronize_sched(). */
|
2010-10-21 05:29:05 +02:00
|
|
|
if (trycount++ < 10)
|
|
|
|
udelay(trycount * num_online_cpus());
|
|
|
|
else {
|
|
|
|
synchronize_sched();
|
|
|
|
return;
|
|
|
|
}
|
2010-11-23 06:36:11 +01:00
|
|
|
|
|
|
|
/* Check to see if someone else did our work for us. */
|
|
|
|
s = atomic_read(&sync_sched_expedited_done);
|
|
|
|
if (UINT_CMP_GE((unsigned)s, (unsigned)firstsnap)) {
|
2010-10-21 05:29:05 +02:00
|
|
|
smp_mb(); /* ensure test happens before caller kfree */
|
|
|
|
return;
|
|
|
|
}
|
2010-11-23 06:36:11 +01:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Refetching sync_sched_expedited_started allows later
|
|
|
|
* callers to piggyback on our grace period. We subtract
|
|
|
|
* 1 to get the same token that the last incrementer got.
|
|
|
|
* We retry after they started, so our grace period works
|
|
|
|
* for them, and they started after our first try, so their
|
|
|
|
* grace period works for us.
|
|
|
|
*/
|
2010-10-21 05:29:05 +02:00
|
|
|
get_online_cpus();
|
2010-11-23 06:36:11 +01:00
|
|
|
snap = atomic_read(&sync_sched_expedited_started) - 1;
|
|
|
|
smp_mb(); /* ensure read is before try_stop_cpus(). */
|
2010-10-21 05:29:05 +02:00
|
|
|
}
|
2010-11-23 06:36:11 +01:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Everyone up to our most recent fetch is covered by our grace
|
|
|
|
* period. Update the counter, but only if our work is still
|
|
|
|
* relevant -- which it won't be if someone who started later
|
|
|
|
* than we did beat us to the punch.
|
|
|
|
*/
|
|
|
|
do {
|
|
|
|
s = atomic_read(&sync_sched_expedited_done);
|
|
|
|
if (UINT_CMP_GE((unsigned)s, (unsigned)snap)) {
|
|
|
|
smp_mb(); /* ensure test happens before caller kfree */
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
} while (atomic_cmpxchg(&sync_sched_expedited_done, s, snap) != s);
|
|
|
|
|
2010-10-21 05:29:05 +02:00
|
|
|
put_online_cpus();
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(synchronize_sched_expedited);
|
|
|
|
|
|
|
|
#endif /* #else #ifndef CONFIG_SMP */
|
|
|
|
|
2010-02-23 02:04:59 +01:00
|
|
|
#if !defined(CONFIG_RCU_FAST_NO_HZ)
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Check to see if any future RCU-related work will need to be done
|
|
|
|
* by the current CPU, even if none need be done immediately, returning
|
|
|
|
* 1 if so. This function is part of the RCU implementation; it is -not-
|
|
|
|
* an exported member of the RCU API.
|
|
|
|
*
|
|
|
|
* Because we have preemptible RCU, just check whether this CPU needs
|
|
|
|
* any flavor of RCU. Do not chew up lots of CPU cycles with preemption
|
|
|
|
* disabled in a most-likely vain attempt to cause RCU not to need this CPU.
|
|
|
|
*/
|
|
|
|
int rcu_needs_cpu(int cpu)
|
|
|
|
{
|
|
|
|
return rcu_needs_cpu_quick_check(cpu);
|
|
|
|
}
|
|
|
|
|
2010-02-27 01:38:56 +01:00
|
|
|
/*
|
|
|
|
* Check to see if we need to continue a callback-flush operations to
|
|
|
|
* allow the last CPU to enter dyntick-idle mode. But fast dyntick-idle
|
|
|
|
* entry is not configured, so we never do need to.
|
|
|
|
*/
|
|
|
|
static void rcu_needs_cpu_flush(void)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2010-02-23 02:04:59 +01:00
|
|
|
#else /* #if !defined(CONFIG_RCU_FAST_NO_HZ) */
|
|
|
|
|
|
|
|
#define RCU_NEEDS_CPU_FLUSHES 5
|
2010-02-27 01:38:56 +01:00
|
|
|
static DEFINE_PER_CPU(int, rcu_dyntick_drain);
|
2010-02-27 01:38:58 +01:00
|
|
|
static DEFINE_PER_CPU(unsigned long, rcu_dyntick_holdoff);
|
2010-02-23 02:04:59 +01:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Check to see if any future RCU-related work will need to be done
|
|
|
|
* by the current CPU, even if none need be done immediately, returning
|
|
|
|
* 1 if so. This function is part of the RCU implementation; it is -not-
|
|
|
|
* an exported member of the RCU API.
|
|
|
|
*
|
|
|
|
* Because we are not supporting preemptible RCU, attempt to accelerate
|
|
|
|
* any current grace periods so that RCU no longer needs this CPU, but
|
|
|
|
* only if all other CPUs are already in dynticks-idle mode. This will
|
|
|
|
* allow the CPU cores to be powered down immediately, as opposed to after
|
|
|
|
* waiting many milliseconds for grace periods to elapse.
|
2010-02-27 01:38:56 +01:00
|
|
|
*
|
|
|
|
* Because it is not legal to invoke rcu_process_callbacks() with irqs
|
|
|
|
* disabled, we do one pass of force_quiescent_state(), then do a
|
2011-02-07 21:47:15 +01:00
|
|
|
* invoke_rcu_cpu_kthread() to cause rcu_process_callbacks() to be invoked
|
|
|
|
* later. The per-cpu rcu_dyntick_drain variable controls the sequencing.
|
2010-02-23 02:04:59 +01:00
|
|
|
*/
|
|
|
|
int rcu_needs_cpu(int cpu)
|
|
|
|
{
|
2010-02-27 01:38:56 +01:00
|
|
|
int c = 0;
|
2010-04-26 06:04:29 +02:00
|
|
|
int snap;
|
2010-02-23 02:04:59 +01:00
|
|
|
int thatcpu;
|
|
|
|
|
2010-02-27 23:53:07 +01:00
|
|
|
/* Check for being in the holdoff period. */
|
|
|
|
if (per_cpu(rcu_dyntick_holdoff, cpu) == jiffies)
|
|
|
|
return rcu_needs_cpu_quick_check(cpu);
|
|
|
|
|
2010-02-23 02:04:59 +01:00
|
|
|
/* Don't bother unless we are the last non-dyntick-idle CPU. */
|
2010-04-26 06:04:29 +02:00
|
|
|
for_each_online_cpu(thatcpu) {
|
|
|
|
if (thatcpu == cpu)
|
|
|
|
continue;
|
rcu: Decrease memory-barrier usage based on semi-formal proof
(Note: this was reverted, and is now being re-applied in pieces, with
this being the fifth and final piece. See below for the reason that
it is now felt to be safe to re-apply this.)
Commit d09b62d fixed grace-period synchronization, but left some smp_mb()
invocations in rcu_process_callbacks() that are no longer needed, but
sheer paranoia prevented them from being removed. This commit removes
them and provides a proof of correctness in their absence. It also adds
a memory barrier to rcu_report_qs_rsp() immediately before the update to
rsp->completed in order to handle the theoretical possibility that the
compiler or CPU might move massive quantities of code into a lock-based
critical section. This also proves that the sheer paranoia was not
entirely unjustified, at least from a theoretical point of view.
In addition, the old dyntick-idle synchronization depended on the fact
that grace periods were many milliseconds in duration, so that it could
be assumed that no dyntick-idle CPU could reorder a memory reference
across an entire grace period. Unfortunately for this design, the
addition of expedited grace periods breaks this assumption, which has
the unfortunate side-effect of requiring atomic operations in the
functions that track dyntick-idle state for RCU. (There is some hope
that the algorithms used in user-level RCU might be applied here, but
some work is required to handle the NMIs that user-space applications
can happily ignore. For the short term, better safe than sorry.)
This proof assumes that neither compiler nor CPU will allow a lock
acquisition and release to be reordered, as doing so can result in
deadlock. The proof is as follows:
1. A given CPU declares a quiescent state under the protection of
its leaf rcu_node's lock.
2. If there is more than one level of rcu_node hierarchy, the
last CPU to declare a quiescent state will also acquire the
->lock of the next rcu_node up in the hierarchy, but only
after releasing the lower level's lock. The acquisition of this
lock clearly cannot occur prior to the acquisition of the leaf
node's lock.
3. Step 2 repeats until we reach the root rcu_node structure.
Please note again that only one lock is held at a time through
this process. The acquisition of the root rcu_node's ->lock
must occur after the release of that of the leaf rcu_node.
4. At this point, we set the ->completed field in the rcu_state
structure in rcu_report_qs_rsp(). However, if the rcu_node
hierarchy contains only one rcu_node, then in theory the code
preceding the quiescent state could leak into the critical
section. We therefore precede the update of ->completed with a
memory barrier. All CPUs will therefore agree that any updates
preceding any report of a quiescent state will have happened
before the update of ->completed.
5. Regardless of whether a new grace period is needed, rcu_start_gp()
will propagate the new value of ->completed to all of the leaf
rcu_node structures, under the protection of each rcu_node's ->lock.
If a new grace period is needed immediately, this propagation
will occur in the same critical section that ->completed was
set in, but courtesy of the memory barrier in #4 above, is still
seen to follow any pre-quiescent-state activity.
6. When a given CPU invokes __rcu_process_gp_end(), it becomes
aware of the end of the old grace period and therefore makes
any RCU callbacks that were waiting on that grace period eligible
for invocation.
If this CPU is the same one that detected the end of the grace
period, and if there is but a single rcu_node in the hierarchy,
we will still be in the single critical section. In this case,
the memory barrier in step #4 guarantees that all callbacks will
be seen to execute after each CPU's quiescent state.
On the other hand, if this is a different CPU, it will acquire
the leaf rcu_node's ->lock, and will again be serialized after
each CPU's quiescent state for the old grace period.
On the strength of this proof, this commit therefore removes the memory
barriers from rcu_process_callbacks() and adds one to rcu_report_qs_rsp().
The effect is to reduce the number of memory barriers by one and to
reduce the frequency of execution from about once per scheduling tick
per CPU to once per grace period.
This was reverted do to hangs found during testing by Yinghai Lu and
Ingo Molnar. Frederic Weisbecker supplied Yinghai with tracing that
located the underlying problem, and Frederic also provided the fix.
The underlying problem was that the HARDIRQ_ENTER() macro from
lib/locking-selftest.c invoked irq_enter(), which in turn invokes
rcu_irq_enter(), but HARDIRQ_EXIT() invoked __irq_exit(), which
does not invoke rcu_irq_exit(). This situation resulted in calls
to rcu_irq_enter() that were not balanced by the required calls to
rcu_irq_exit(). Therefore, after these locking selftests completed,
RCU's dyntick-idle nesting count was a large number (for example,
72), which caused RCU to to conclude that the affected CPU was not in
dyntick-idle mode when in fact it was.
RCU would therefore incorrectly wait for this dyntick-idle CPU, resulting
in hangs.
In contrast, with Frederic's patch, which replaces the irq_enter()
in HARDIRQ_ENTER() with an __irq_enter(), these tests don't ever call
either rcu_irq_enter() or rcu_irq_exit(), which works because the CPU
running the test is already marked as not being in dyntick-idle mode.
This means that the rcu_irq_enter() and rcu_irq_exit() calls and RCU
then has no problem working out which CPUs are in dyntick-idle mode and
which are not.
The reason that the imbalance was not noticed before the barrier patch
was applied is that the old implementation of rcu_enter_nohz() ignored
the nesting depth. This could still result in delays, but much shorter
ones. Whenever there was a delay, RCU would IPI the CPU with the
unbalanced nesting level, which would eventually result in rcu_enter_nohz()
being called, which in turn would force RCU to see that the CPU was in
dyntick-idle mode.
The reason that very few people noticed the problem is that the mismatched
irq_enter() vs. __irq_exit() occured only when the kernel was built with
CONFIG_DEBUG_LOCKING_API_SELFTESTS.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
2010-09-07 19:38:22 +02:00
|
|
|
snap = atomic_add_return(0, &per_cpu(rcu_dynticks,
|
|
|
|
thatcpu).dynticks);
|
2010-04-26 06:04:29 +02:00
|
|
|
smp_mb(); /* Order sampling of snap with end of grace period. */
|
rcu: Decrease memory-barrier usage based on semi-formal proof
(Note: this was reverted, and is now being re-applied in pieces, with
this being the fifth and final piece. See below for the reason that
it is now felt to be safe to re-apply this.)
Commit d09b62d fixed grace-period synchronization, but left some smp_mb()
invocations in rcu_process_callbacks() that are no longer needed, but
sheer paranoia prevented them from being removed. This commit removes
them and provides a proof of correctness in their absence. It also adds
a memory barrier to rcu_report_qs_rsp() immediately before the update to
rsp->completed in order to handle the theoretical possibility that the
compiler or CPU might move massive quantities of code into a lock-based
critical section. This also proves that the sheer paranoia was not
entirely unjustified, at least from a theoretical point of view.
In addition, the old dyntick-idle synchronization depended on the fact
that grace periods were many milliseconds in duration, so that it could
be assumed that no dyntick-idle CPU could reorder a memory reference
across an entire grace period. Unfortunately for this design, the
addition of expedited grace periods breaks this assumption, which has
the unfortunate side-effect of requiring atomic operations in the
functions that track dyntick-idle state for RCU. (There is some hope
that the algorithms used in user-level RCU might be applied here, but
some work is required to handle the NMIs that user-space applications
can happily ignore. For the short term, better safe than sorry.)
This proof assumes that neither compiler nor CPU will allow a lock
acquisition and release to be reordered, as doing so can result in
deadlock. The proof is as follows:
1. A given CPU declares a quiescent state under the protection of
its leaf rcu_node's lock.
2. If there is more than one level of rcu_node hierarchy, the
last CPU to declare a quiescent state will also acquire the
->lock of the next rcu_node up in the hierarchy, but only
after releasing the lower level's lock. The acquisition of this
lock clearly cannot occur prior to the acquisition of the leaf
node's lock.
3. Step 2 repeats until we reach the root rcu_node structure.
Please note again that only one lock is held at a time through
this process. The acquisition of the root rcu_node's ->lock
must occur after the release of that of the leaf rcu_node.
4. At this point, we set the ->completed field in the rcu_state
structure in rcu_report_qs_rsp(). However, if the rcu_node
hierarchy contains only one rcu_node, then in theory the code
preceding the quiescent state could leak into the critical
section. We therefore precede the update of ->completed with a
memory barrier. All CPUs will therefore agree that any updates
preceding any report of a quiescent state will have happened
before the update of ->completed.
5. Regardless of whether a new grace period is needed, rcu_start_gp()
will propagate the new value of ->completed to all of the leaf
rcu_node structures, under the protection of each rcu_node's ->lock.
If a new grace period is needed immediately, this propagation
will occur in the same critical section that ->completed was
set in, but courtesy of the memory barrier in #4 above, is still
seen to follow any pre-quiescent-state activity.
6. When a given CPU invokes __rcu_process_gp_end(), it becomes
aware of the end of the old grace period and therefore makes
any RCU callbacks that were waiting on that grace period eligible
for invocation.
If this CPU is the same one that detected the end of the grace
period, and if there is but a single rcu_node in the hierarchy,
we will still be in the single critical section. In this case,
the memory barrier in step #4 guarantees that all callbacks will
be seen to execute after each CPU's quiescent state.
On the other hand, if this is a different CPU, it will acquire
the leaf rcu_node's ->lock, and will again be serialized after
each CPU's quiescent state for the old grace period.
On the strength of this proof, this commit therefore removes the memory
barriers from rcu_process_callbacks() and adds one to rcu_report_qs_rsp().
The effect is to reduce the number of memory barriers by one and to
reduce the frequency of execution from about once per scheduling tick
per CPU to once per grace period.
This was reverted do to hangs found during testing by Yinghai Lu and
Ingo Molnar. Frederic Weisbecker supplied Yinghai with tracing that
located the underlying problem, and Frederic also provided the fix.
The underlying problem was that the HARDIRQ_ENTER() macro from
lib/locking-selftest.c invoked irq_enter(), which in turn invokes
rcu_irq_enter(), but HARDIRQ_EXIT() invoked __irq_exit(), which
does not invoke rcu_irq_exit(). This situation resulted in calls
to rcu_irq_enter() that were not balanced by the required calls to
rcu_irq_exit(). Therefore, after these locking selftests completed,
RCU's dyntick-idle nesting count was a large number (for example,
72), which caused RCU to to conclude that the affected CPU was not in
dyntick-idle mode when in fact it was.
RCU would therefore incorrectly wait for this dyntick-idle CPU, resulting
in hangs.
In contrast, with Frederic's patch, which replaces the irq_enter()
in HARDIRQ_ENTER() with an __irq_enter(), these tests don't ever call
either rcu_irq_enter() or rcu_irq_exit(), which works because the CPU
running the test is already marked as not being in dyntick-idle mode.
This means that the rcu_irq_enter() and rcu_irq_exit() calls and RCU
then has no problem working out which CPUs are in dyntick-idle mode and
which are not.
The reason that the imbalance was not noticed before the barrier patch
was applied is that the old implementation of rcu_enter_nohz() ignored
the nesting depth. This could still result in delays, but much shorter
ones. Whenever there was a delay, RCU would IPI the CPU with the
unbalanced nesting level, which would eventually result in rcu_enter_nohz()
being called, which in turn would force RCU to see that the CPU was in
dyntick-idle mode.
The reason that very few people noticed the problem is that the mismatched
irq_enter() vs. __irq_exit() occured only when the kernel was built with
CONFIG_DEBUG_LOCKING_API_SELFTESTS.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
2010-09-07 19:38:22 +02:00
|
|
|
if ((snap & 0x1) != 0) {
|
2010-02-27 01:38:56 +01:00
|
|
|
per_cpu(rcu_dyntick_drain, cpu) = 0;
|
2010-02-27 01:38:58 +01:00
|
|
|
per_cpu(rcu_dyntick_holdoff, cpu) = jiffies - 1;
|
2010-02-23 02:04:59 +01:00
|
|
|
return rcu_needs_cpu_quick_check(cpu);
|
|
|
|
}
|
2010-04-26 06:04:29 +02:00
|
|
|
}
|
2010-02-27 01:38:56 +01:00
|
|
|
|
|
|
|
/* Check and update the rcu_dyntick_drain sequencing. */
|
|
|
|
if (per_cpu(rcu_dyntick_drain, cpu) <= 0) {
|
|
|
|
/* First time through, initialize the counter. */
|
|
|
|
per_cpu(rcu_dyntick_drain, cpu) = RCU_NEEDS_CPU_FLUSHES;
|
|
|
|
} else if (--per_cpu(rcu_dyntick_drain, cpu) <= 0) {
|
|
|
|
/* We have hit the limit, so time to give up. */
|
2010-02-27 01:38:58 +01:00
|
|
|
per_cpu(rcu_dyntick_holdoff, cpu) = jiffies;
|
2010-02-27 01:38:56 +01:00
|
|
|
return rcu_needs_cpu_quick_check(cpu);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Do one step pushing remaining RCU callbacks through. */
|
|
|
|
if (per_cpu(rcu_sched_data, cpu).nxtlist) {
|
|
|
|
rcu_sched_qs(cpu);
|
|
|
|
force_quiescent_state(&rcu_sched_state, 0);
|
|
|
|
c = c || per_cpu(rcu_sched_data, cpu).nxtlist;
|
|
|
|
}
|
|
|
|
if (per_cpu(rcu_bh_data, cpu).nxtlist) {
|
|
|
|
rcu_bh_qs(cpu);
|
|
|
|
force_quiescent_state(&rcu_bh_state, 0);
|
|
|
|
c = c || per_cpu(rcu_bh_data, cpu).nxtlist;
|
2010-02-23 02:04:59 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
/* If RCU callbacks are still pending, RCU still needs this CPU. */
|
2010-02-27 23:53:07 +01:00
|
|
|
if (c)
|
2011-02-07 21:47:15 +01:00
|
|
|
invoke_rcu_cpu_kthread();
|
2010-02-23 02:04:59 +01:00
|
|
|
return c;
|
|
|
|
}
|
|
|
|
|
2010-02-27 01:38:56 +01:00
|
|
|
/*
|
|
|
|
* Check to see if we need to continue a callback-flush operations to
|
|
|
|
* allow the last CPU to enter dyntick-idle mode.
|
|
|
|
*/
|
|
|
|
static void rcu_needs_cpu_flush(void)
|
|
|
|
{
|
|
|
|
int cpu = smp_processor_id();
|
2010-02-27 01:38:58 +01:00
|
|
|
unsigned long flags;
|
2010-02-27 01:38:56 +01:00
|
|
|
|
|
|
|
if (per_cpu(rcu_dyntick_drain, cpu) <= 0)
|
|
|
|
return;
|
2010-02-27 01:38:58 +01:00
|
|
|
local_irq_save(flags);
|
2010-02-27 01:38:56 +01:00
|
|
|
(void)rcu_needs_cpu(cpu);
|
2010-02-27 01:38:58 +01:00
|
|
|
local_irq_restore(flags);
|
2010-02-27 01:38:56 +01:00
|
|
|
}
|
|
|
|
|
2010-02-23 02:04:59 +01:00
|
|
|
#endif /* #else #if !defined(CONFIG_RCU_FAST_NO_HZ) */
|