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>
2009-08-23 04:56:52 +08:00
|
|
|
/*
|
|
|
|
* Read-Copy Update mechanism for mutual exclusion (tree-based version)
|
|
|
|
* Internal non-public definitions that provide either classic
|
|
|
|
* or preemptable semantics.
|
|
|
|
*
|
|
|
|
* 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>
|
|
|
|
*/
|
|
|
|
|
2009-12-03 04:10:15 +08:00
|
|
|
#include <linux/delay.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>
2009-08-23 04:56:52 +08:00
|
|
|
|
|
|
|
#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);
|
|
|
|
|
2009-12-03 04:10:15 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
/*
|
|
|
|
* Tell them what RCU they are running.
|
|
|
|
*/
|
2009-11-12 03:28:06 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
{
|
|
|
|
printk(KERN_INFO
|
|
|
|
"Experimental preemptable hierarchical RCU implementation.\n");
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Record a preemptable-RCU quiescent state for the specified CPU. Note
|
|
|
|
* 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.
|
|
|
|
*/
|
2009-09-14 00:15:10 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
{
|
|
|
|
struct rcu_data *rdp = &per_cpu(rcu_preempt_data, cpu);
|
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>
2009-11-11 05:37:22 +08:00
|
|
|
rdp->passed_quiesc_completed = rdp->gpnum - 1;
|
2009-09-14 00:15:10 +08:00
|
|
|
barrier();
|
|
|
|
rdp->passed_quiesc = 1;
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2009-09-14 00:15:10 +08:00
|
|
|
* 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
|
|
|
|
* record that fact, so we enqueue the task on the appropriate entry
|
|
|
|
* of the blocked_tasks[] array. 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 blocked_tasks[] entry indexed by the low-order bit of
|
|
|
|
* rnp->gpnum empties.
|
|
|
|
*
|
|
|
|
* 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>
2009-08-23 04:56:52 +08:00
|
|
|
*/
|
2009-09-14 00:15:10 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
{
|
|
|
|
struct task_struct *t = current;
|
2009-09-14 00:15:10 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
int phase;
|
|
|
|
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. */
|
|
|
|
rdp = rcu_preempt_state.rda[cpu];
|
|
|
|
rnp = rdp->mynode;
|
2009-09-14 00:15:10 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BLOCKED;
|
2009-08-28 06:00:12 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
|
|
|
|
/*
|
|
|
|
* 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
|
|
|
|
* cannot end.
|
2009-09-14 00:15:09 +08:00
|
|
|
*
|
|
|
|
* 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>
2009-08-23 04:56:52 +08:00
|
|
|
*/
|
2009-09-14 00:15:09 +08:00
|
|
|
WARN_ON_ONCE((rdp->grpmask & rnp->qsmaskinit) == 0);
|
2009-09-19 00:50:18 +08:00
|
|
|
WARN_ON_ONCE(!list_empty(&t->rcu_node_entry));
|
|
|
|
phase = (rnp->gpnum + !(rnp->qsmask & rdp->grpmask)) & 0x1;
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
list_add(&t->rcu_node_entry, &rnp->blocked_tasks[phase]);
|
2009-09-14 00:15:10 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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.
|
|
|
|
*/
|
2009-09-14 00:15:10 +08:00
|
|
|
rcu_preempt_qs(cpu);
|
2009-09-19 00:50:18 +08:00
|
|
|
local_irq_save(flags);
|
2009-09-14 00:15:10 +08:00
|
|
|
t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS;
|
2009-09-19 00:50:18 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Tree-preemptable RCU implementation for rcu_read_lock().
|
|
|
|
* Just increment ->rcu_read_lock_nesting, shared state will be updated
|
|
|
|
* if we block.
|
|
|
|
*/
|
|
|
|
void __rcu_read_lock(void)
|
|
|
|
{
|
|
|
|
ACCESS_ONCE(current->rcu_read_lock_nesting)++;
|
|
|
|
barrier(); /* needed if we ever invoke rcu_read_lock in rcutree.c */
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(__rcu_read_lock);
|
|
|
|
|
2009-09-24 00:50:41 +08:00
|
|
|
/*
|
|
|
|
* 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.
|
|
|
|
*/
|
|
|
|
static int rcu_preempted_readers(struct rcu_node *rnp)
|
|
|
|
{
|
2009-12-03 04:10:15 +08:00
|
|
|
int phase = rnp->gpnum & 0x1;
|
|
|
|
|
|
|
|
return !list_empty(&rnp->blocked_tasks[phase]) ||
|
|
|
|
!list_empty(&rnp->blocked_tasks[phase + 2]);
|
2009-09-24 00:50:41 +08:00
|
|
|
}
|
|
|
|
|
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>
2009-11-23 00:53:48 +08:00
|
|
|
/*
|
|
|
|
* 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.
|
|
|
|
*/
|
2009-12-03 04:10:13 +08:00
|
|
|
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>
2009-11-23 00:53:48 +08:00
|
|
|
__releases(rnp->lock)
|
|
|
|
{
|
|
|
|
unsigned long mask;
|
|
|
|
struct rcu_node *rnp_p;
|
|
|
|
|
|
|
|
if (rnp->qsmask != 0 || rcu_preempted_readers(rnp)) {
|
|
|
|
spin_unlock_irqrestore(&rnp->lock, flags);
|
|
|
|
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.
|
|
|
|
*/
|
2009-12-03 04:10:13 +08:00
|
|
|
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>
2009-11-23 00:53:48 +08:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Report up the rest of the hierarchy. */
|
|
|
|
mask = rnp->grpmask;
|
|
|
|
spin_unlock(&rnp->lock); /* irqs remain disabled. */
|
|
|
|
spin_lock(&rnp_p->lock); /* irqs already disabled. */
|
2009-12-03 04:10:13 +08:00
|
|
|
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>
2009-11-23 00:53:48 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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>
2009-08-23 04:56:52 +08:00
|
|
|
static void rcu_read_unlock_special(struct task_struct *t)
|
|
|
|
{
|
|
|
|
int empty;
|
2009-12-03 04:10:15 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
unsigned long flags;
|
|
|
|
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) {
|
|
|
|
t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS;
|
2009-09-14 00:15:10 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/* 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;
|
|
|
|
|
2009-08-28 05:58:16 +08:00
|
|
|
/*
|
|
|
|
* 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 06:00:12 +08:00
|
|
|
rnp = t->rcu_blocked_node;
|
2009-09-19 00:50:18 +08:00
|
|
|
spin_lock(&rnp->lock); /* irqs already disabled. */
|
2009-08-28 06:00:12 +08:00
|
|
|
if (rnp == t->rcu_blocked_node)
|
2009-08-28 05:58:16 +08:00
|
|
|
break;
|
2009-09-19 00:50:18 +08:00
|
|
|
spin_unlock(&rnp->lock); /* irqs remain disabled. */
|
2009-08-28 05:58:16 +08:00
|
|
|
}
|
2009-09-24 00:50:41 +08:00
|
|
|
empty = !rcu_preempted_readers(rnp);
|
2009-12-03 04:10:15 +08:00
|
|
|
empty_exp = !rcu_preempted_readers_exp(rnp);
|
|
|
|
smp_mb(); /* ensure expedited fastpath sees end of RCU c-s. */
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
list_del_init(&t->rcu_node_entry);
|
2009-08-28 05:58:16 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
|
|
|
|
/*
|
|
|
|
* If this was the last task on the current list, and if
|
|
|
|
* we aren't waiting on any CPUs, report the quiescent state.
|
2009-12-03 04:10:13 +08:00
|
|
|
* 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>
2009-08-23 04:56:52 +08:00
|
|
|
*/
|
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>
2009-11-23 00:53:48 +08:00
|
|
|
if (empty)
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
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>
2009-11-23 00:53:48 +08:00
|
|
|
else
|
2009-12-03 04:10:13 +08:00
|
|
|
rcu_report_unblock_qs_rnp(rnp, flags);
|
2009-12-03 04:10:15 +08:00
|
|
|
|
|
|
|
/*
|
|
|
|
* 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>
2009-11-23 00:53:48 +08:00
|
|
|
} 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>
2009-08-23 04:56:52 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Tree-preemptable RCU implementation for rcu_read_unlock().
|
|
|
|
* 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 */
|
|
|
|
if (--ACCESS_ONCE(t->rcu_read_lock_nesting) == 0 &&
|
|
|
|
unlikely(ACCESS_ONCE(t->rcu_read_unlock_special)))
|
|
|
|
rcu_read_unlock_special(t);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(__rcu_read_unlock);
|
|
|
|
|
|
|
|
#ifdef CONFIG_RCU_CPU_STALL_DETECTOR
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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)
|
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
struct list_head *lp;
|
2009-09-24 00:50:41 +08:00
|
|
|
int phase;
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
struct task_struct *t;
|
|
|
|
|
2009-09-24 00:50:41 +08:00
|
|
|
if (rcu_preempted_readers(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>
2009-08-23 04:56:52 +08:00
|
|
|
spin_lock_irqsave(&rnp->lock, flags);
|
2009-09-24 00:50:41 +08:00
|
|
|
phase = rnp->gpnum & 0x1;
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
lp = &rnp->blocked_tasks[phase];
|
|
|
|
list_for_each_entry(t, lp, rcu_node_entry)
|
|
|
|
printk(" P%d", t->pid);
|
|
|
|
spin_unlock_irqrestore(&rnp->lock, flags);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */
|
|
|
|
|
2009-09-14 00:15:09 +08:00
|
|
|
/*
|
|
|
|
* 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.
|
|
|
|
*/
|
|
|
|
static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp)
|
|
|
|
{
|
2009-09-24 00:50:41 +08:00
|
|
|
WARN_ON_ONCE(rcu_preempted_readers(rnp));
|
2009-09-19 00:50:17 +08:00
|
|
|
WARN_ON_ONCE(rnp->qsmask);
|
2009-09-14 00:15:09 +08:00
|
|
|
}
|
|
|
|
|
2009-08-25 00:42:01 +08:00
|
|
|
#ifdef CONFIG_HOTPLUG_CPU
|
|
|
|
|
2009-08-28 05:58:16 +08:00
|
|
|
/*
|
|
|
|
* 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>
2009-11-23 00:53:48 +08:00
|
|
|
* Returns true if there were tasks blocking the current RCU grace
|
|
|
|
* period.
|
2009-08-28 05:58:16 +08:00
|
|
|
*
|
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>
2009-10-16 00:26:14 +08:00
|
|
|
* Returns 1 if there was previously a task blocking the current grace
|
|
|
|
* period on the specified rcu_node structure.
|
|
|
|
*
|
2009-08-28 05:58:16 +08:00
|
|
|
* 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>
2009-10-16 00:26:14 +08:00
|
|
|
static int rcu_preempt_offline_tasks(struct rcu_state *rsp,
|
|
|
|
struct rcu_node *rnp,
|
|
|
|
struct rcu_data *rdp)
|
2009-08-28 05:58:16 +08:00
|
|
|
{
|
|
|
|
int i;
|
|
|
|
struct list_head *lp;
|
|
|
|
struct list_head *lp_root;
|
2009-12-03 04:10:15 +08:00
|
|
|
int retval = 0;
|
2009-08-28 05:58:16 +08:00
|
|
|
struct rcu_node *rnp_root = rcu_get_root(rsp);
|
|
|
|
struct task_struct *tp;
|
|
|
|
|
2009-08-28 06:00:12 +08:00
|
|
|
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>
2009-10-16 00:26:14 +08:00
|
|
|
return 0; /* Shouldn't happen: at least one CPU online. */
|
2009-08-28 06:00:12 +08:00
|
|
|
}
|
2009-09-19 00:50:17 +08:00
|
|
|
WARN_ON_ONCE(rnp != rdp->mynode &&
|
|
|
|
(!list_empty(&rnp->blocked_tasks[0]) ||
|
2009-12-03 04:10:15 +08:00
|
|
|
!list_empty(&rnp->blocked_tasks[1]) ||
|
|
|
|
!list_empty(&rnp->blocked_tasks[2]) ||
|
|
|
|
!list_empty(&rnp->blocked_tasks[3])));
|
2009-08-28 05:58:16 +08:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Move tasks up to root rcu_node. Rely on the fact that the
|
|
|
|
* root rcu_node can be at most one ahead of the rest of the
|
|
|
|
* rcu_nodes in terms of gp_num value. This fact allows us to
|
|
|
|
* move the blocked_tasks[] array directly, element by element.
|
|
|
|
*/
|
2009-12-03 04:10:15 +08:00
|
|
|
if (rcu_preempted_readers(rnp))
|
|
|
|
retval |= RCU_OFL_TASKS_NORM_GP;
|
|
|
|
if (rcu_preempted_readers_exp(rnp))
|
|
|
|
retval |= RCU_OFL_TASKS_EXP_GP;
|
|
|
|
for (i = 0; i < 4; i++) {
|
2009-08-28 05:58:16 +08:00
|
|
|
lp = &rnp->blocked_tasks[i];
|
|
|
|
lp_root = &rnp_root->blocked_tasks[i];
|
|
|
|
while (!list_empty(lp)) {
|
|
|
|
tp = list_entry(lp->next, typeof(*tp), rcu_node_entry);
|
|
|
|
spin_lock(&rnp_root->lock); /* irqs already disabled */
|
|
|
|
list_del(&tp->rcu_node_entry);
|
|
|
|
tp->rcu_blocked_node = rnp_root;
|
|
|
|
list_add(&tp->rcu_node_entry, lp_root);
|
|
|
|
spin_unlock(&rnp_root->lock); /* irqs remain 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>
2009-10-16 00:26:14 +08:00
|
|
|
return retval;
|
2009-08-28 05:58:16 +08:00
|
|
|
}
|
|
|
|
|
2009-08-25 00:42:01 +08:00
|
|
|
/*
|
|
|
|
* Do CPU-offline processing for preemptable RCU.
|
|
|
|
*/
|
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
/*
|
|
|
|
* 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) {
|
2009-09-14 00:15:10 +08:00
|
|
|
t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS;
|
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
return;
|
|
|
|
}
|
2009-09-19 01:28:19 +08:00
|
|
|
if (per_cpu(rcu_preempt_data, cpu).qs_pending)
|
2009-09-14 00:15:10 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Process callbacks for preemptable RCU.
|
|
|
|
*/
|
|
|
|
static void rcu_preempt_process_callbacks(void)
|
|
|
|
{
|
|
|
|
__rcu_process_callbacks(&rcu_preempt_state,
|
|
|
|
&__get_cpu_var(rcu_preempt_data));
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Queue a preemptable-RCU callback for invocation after a grace period.
|
|
|
|
*/
|
|
|
|
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);
|
|
|
|
|
2009-11-23 00:53:50 +08:00
|
|
|
/**
|
|
|
|
* 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
|
|
|
|
* read-side critical sections have completed. RCU read-side critical
|
|
|
|
* sections are delimited by rcu_read_lock() and rcu_read_unlock(),
|
|
|
|
* and may be nested.
|
|
|
|
*/
|
|
|
|
void synchronize_rcu(void)
|
|
|
|
{
|
|
|
|
struct rcu_synchronize rcu;
|
|
|
|
|
|
|
|
if (!rcu_scheduler_active)
|
|
|
|
return;
|
|
|
|
|
|
|
|
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);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(synchronize_rcu);
|
|
|
|
|
2009-12-03 04:10:15 +08:00
|
|
|
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)
|
|
|
|
{
|
|
|
|
return !list_empty(&rnp->blocked_tasks[2]) ||
|
|
|
|
!list_empty(&rnp->blocked_tasks[3]);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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;
|
|
|
|
|
|
|
|
spin_lock_irqsave(&rnp->lock, flags);
|
|
|
|
for (;;) {
|
|
|
|
if (!sync_rcu_preempt_exp_done(rnp))
|
|
|
|
break;
|
|
|
|
if (rnp->parent == NULL) {
|
|
|
|
wake_up(&sync_rcu_preempt_exp_wq);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
mask = rnp->grpmask;
|
|
|
|
spin_unlock(&rnp->lock); /* irqs remain disabled */
|
|
|
|
rnp = rnp->parent;
|
|
|
|
spin_lock(&rnp->lock); /* irqs already disabled */
|
|
|
|
rnp->expmask &= ~mask;
|
|
|
|
}
|
|
|
|
spin_unlock_irqrestore(&rnp->lock, flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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)
|
|
|
|
{
|
|
|
|
int must_wait;
|
|
|
|
|
|
|
|
spin_lock(&rnp->lock); /* irqs already disabled */
|
|
|
|
list_splice_init(&rnp->blocked_tasks[0], &rnp->blocked_tasks[2]);
|
|
|
|
list_splice_init(&rnp->blocked_tasks[1], &rnp->blocked_tasks[3]);
|
|
|
|
must_wait = rcu_preempted_readers_exp(rnp);
|
|
|
|
spin_unlock(&rnp->lock); /* irqs remain disabled */
|
|
|
|
if (!must_wait)
|
|
|
|
rcu_report_exp_rnp(rsp, rnp);
|
|
|
|
}
|
|
|
|
|
2009-10-15 01:15:56 +08:00
|
|
|
/*
|
2009-12-03 04:10:15 +08:00
|
|
|
* 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
|
|
|
|
* the ->blocked_tasks[] lists, move all entries from the first set of
|
|
|
|
* ->blocked_tasks[] lists to the second set, and finally wait for this
|
|
|
|
* second set to drain.
|
2009-10-15 01:15:56 +08:00
|
|
|
*/
|
|
|
|
void synchronize_rcu_expedited(void)
|
|
|
|
{
|
2009-12-03 04:10:15 +08:00
|
|
|
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. */
|
|
|
|
|
|
|
|
/* force all RCU readers onto blocked_tasks[]. */
|
|
|
|
synchronize_sched_expedited();
|
|
|
|
|
|
|
|
spin_lock_irqsave(&rsp->onofflock, flags);
|
|
|
|
|
|
|
|
/* Initialize ->expmask for all non-leaf rcu_node structures. */
|
|
|
|
rcu_for_each_nonleaf_node_breadth_first(rsp, rnp) {
|
|
|
|
spin_lock(&rnp->lock); /* irqs already disabled. */
|
|
|
|
rnp->expmask = rnp->qsmaskinit;
|
|
|
|
spin_unlock(&rnp->lock); /* irqs remain disabled. */
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Snapshot current state of ->blocked_tasks[] lists. */
|
|
|
|
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));
|
|
|
|
|
|
|
|
spin_unlock_irqrestore(&rsp->onofflock, flags);
|
|
|
|
|
|
|
|
/* Wait for snapshotted ->blocked_tasks[] lists to drain. */
|
|
|
|
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. */
|
2009-10-15 01:15:56 +08:00
|
|
|
}
|
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
/*
|
|
|
|
* Check to see if there is any immediate preemptable-RCU-related work
|
|
|
|
* to be done.
|
|
|
|
*/
|
|
|
|
static int rcu_preempt_pending(int cpu)
|
|
|
|
{
|
|
|
|
return __rcu_pending(&rcu_preempt_state,
|
|
|
|
&per_cpu(rcu_preempt_data, cpu));
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Does preemptable RCU need the CPU to stay out of dynticks mode?
|
|
|
|
*/
|
|
|
|
static int rcu_preempt_needs_cpu(int cpu)
|
|
|
|
{
|
|
|
|
return !!per_cpu(rcu_preempt_data, cpu).nxtlist;
|
|
|
|
}
|
|
|
|
|
2009-10-07 12:48:17 +08:00
|
|
|
/**
|
|
|
|
* 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>
2009-08-23 04:56:52 +08:00
|
|
|
/*
|
|
|
|
* Initialize preemptable RCU's per-CPU data.
|
|
|
|
*/
|
|
|
|
static void __cpuinit rcu_preempt_init_percpu_data(int cpu)
|
|
|
|
{
|
|
|
|
rcu_init_percpu_data(cpu, &rcu_preempt_state, 1);
|
|
|
|
}
|
|
|
|
|
2009-10-07 12:48:17 +08:00
|
|
|
/*
|
|
|
|
* Move preemptable RCU's callbacks to ->orphan_cbs_list.
|
|
|
|
*/
|
|
|
|
static void rcu_preempt_send_cbs_to_orphanage(void)
|
|
|
|
{
|
|
|
|
rcu_send_cbs_to_orphanage(&rcu_preempt_state);
|
|
|
|
}
|
|
|
|
|
2009-09-24 00:50:42 +08:00
|
|
|
/*
|
|
|
|
* Initialize preemptable RCU's state structures.
|
|
|
|
*/
|
|
|
|
static void __init __rcu_init_preempt(void)
|
|
|
|
{
|
|
|
|
RCU_INIT_FLAVOR(&rcu_preempt_state, rcu_preempt_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>
2009-08-23 04:56:52 +08:00
|
|
|
/*
|
|
|
|
* Check for a task exiting while in a preemptable-RCU read-side
|
|
|
|
* 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;
|
|
|
|
rcu_read_unlock();
|
|
|
|
}
|
|
|
|
|
|
|
|
#else /* #ifdef CONFIG_TREE_PREEMPT_RCU */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Tell them what RCU they are running.
|
|
|
|
*/
|
2009-11-12 03:28:06 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
{
|
|
|
|
printk(KERN_INFO "Hierarchical RCU implementation.\n");
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Because preemptable RCU does not exist, we never have to check for
|
|
|
|
* CPUs being in quiescent states.
|
|
|
|
*/
|
2009-09-14 00:15:10 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2009-09-24 00:50:41 +08:00
|
|
|
/*
|
|
|
|
* Because preemptable RCU does not exist, there are never any preempted
|
|
|
|
* RCU readers.
|
|
|
|
*/
|
|
|
|
static int rcu_preempted_readers(struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
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>
2009-11-23 00:53:48 +08:00
|
|
|
#ifdef CONFIG_HOTPLUG_CPU
|
|
|
|
|
|
|
|
/* Because preemptible RCU does not exist, no quieting of tasks. */
|
2009-12-03 04:10:13 +08:00
|
|
|
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>
2009-11-23 00:53:48 +08:00
|
|
|
{
|
|
|
|
spin_unlock_irqrestore(&rnp->lock, flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
#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>
2009-08-23 04:56:52 +08:00
|
|
|
#ifdef CONFIG_RCU_CPU_STALL_DETECTOR
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Because preemptable RCU does not exist, we never have to check for
|
|
|
|
* tasks blocked within RCU read-side critical sections.
|
|
|
|
*/
|
|
|
|
static void rcu_print_task_stall(struct rcu_node *rnp)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */
|
|
|
|
|
2009-09-14 00:15:09 +08:00
|
|
|
/*
|
|
|
|
* Because there is no preemptable RCU, there can be no readers blocked,
|
2009-09-19 00:50:19 +08:00
|
|
|
* so there is no need to check for blocked tasks. So check only for
|
|
|
|
* bogus qsmask values.
|
2009-09-14 00:15:09 +08:00
|
|
|
*/
|
|
|
|
static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp)
|
|
|
|
{
|
2009-09-19 00:50:19 +08:00
|
|
|
WARN_ON_ONCE(rnp->qsmask);
|
2009-09-14 00:15:09 +08:00
|
|
|
}
|
|
|
|
|
2009-08-25 00:42:01 +08:00
|
|
|
#ifdef CONFIG_HOTPLUG_CPU
|
|
|
|
|
2009-08-28 05:58:16 +08:00
|
|
|
/*
|
|
|
|
* Because preemptable 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>
2009-10-16 00:26:14 +08:00
|
|
|
* tasks that were blocked within RCU read-side critical sections, and
|
|
|
|
* such non-existent tasks cannot possibly have been blocking the current
|
|
|
|
* grace period.
|
2009-08-28 05:58:16 +08:00
|
|
|
*/
|
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>
2009-10-16 00:26:14 +08:00
|
|
|
static int rcu_preempt_offline_tasks(struct rcu_state *rsp,
|
|
|
|
struct rcu_node *rnp,
|
|
|
|
struct rcu_data *rdp)
|
2009-08-28 05:58:16 +08:00
|
|
|
{
|
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>
2009-10-16 00:26:14 +08:00
|
|
|
return 0;
|
2009-08-28 05:58:16 +08:00
|
|
|
}
|
|
|
|
|
2009-08-25 00:42:01 +08:00
|
|
|
/*
|
|
|
|
* Because preemptable RCU does not exist, it never needs CPU-offline
|
|
|
|
* 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>
2009-08-23 04:56:52 +08:00
|
|
|
/*
|
|
|
|
* Because preemptable RCU does not exist, it never has any callbacks
|
|
|
|
* to check.
|
|
|
|
*/
|
2009-09-24 00:50:42 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Because preemptable RCU does not exist, it never has any callbacks
|
|
|
|
* to process.
|
|
|
|
*/
|
2009-09-24 00:50:42 +08:00
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* In classic RCU, call_rcu() is just call_rcu_sched().
|
|
|
|
*/
|
|
|
|
void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
|
|
|
|
{
|
|
|
|
call_rcu_sched(head, func);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(call_rcu);
|
|
|
|
|
2009-10-15 01:15:56 +08:00
|
|
|
/*
|
|
|
|
* Wait for an rcu-preempt grace period, but make it happen quickly.
|
|
|
|
* But because preemptable RCU does not exist, map to rcu-sched.
|
|
|
|
*/
|
|
|
|
void synchronize_rcu_expedited(void)
|
|
|
|
{
|
|
|
|
synchronize_sched_expedited();
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(synchronize_rcu_expedited);
|
|
|
|
|
2009-12-03 04:10:15 +08:00
|
|
|
#ifdef CONFIG_HOTPLUG_CPU
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Because preemptable RCU does not exist, there is never any need to
|
|
|
|
* 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>
2009-08-23 04:56:52 +08:00
|
|
|
/*
|
|
|
|
* Because preemptable RCU does not exist, it never has any work to do.
|
|
|
|
*/
|
|
|
|
static int rcu_preempt_pending(int cpu)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Because preemptable RCU does not exist, it never needs any CPU.
|
|
|
|
*/
|
|
|
|
static int rcu_preempt_needs_cpu(int cpu)
|
|
|
|
{
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-10-07 12:48:17 +08:00
|
|
|
/*
|
|
|
|
* Because preemptable RCU does not exist, rcu_barrier() is just
|
|
|
|
* 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>
2009-08-23 04:56:52 +08:00
|
|
|
/*
|
|
|
|
* Because preemptable RCU does not exist, there is no per-CPU
|
|
|
|
* data to initialize.
|
|
|
|
*/
|
|
|
|
static void __cpuinit rcu_preempt_init_percpu_data(int cpu)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2009-10-07 12:48:17 +08:00
|
|
|
/*
|
|
|
|
* Because there is no preemptable RCU, there are no callbacks to move.
|
|
|
|
*/
|
|
|
|
static void rcu_preempt_send_cbs_to_orphanage(void)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2009-09-24 00:50:42 +08:00
|
|
|
/*
|
|
|
|
* Because preemptable RCU does not exist, it need not be initialized.
|
|
|
|
*/
|
|
|
|
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>
2009-08-23 04:56:52 +08:00
|
|
|
#endif /* #else #ifdef CONFIG_TREE_PREEMPT_RCU */
|