2009-08-22 16:56:45 -04:00
|
|
|
/*
|
|
|
|
* Read-Copy Update mechanism for mutual exclusion (tree-based version)
|
|
|
|
* Internal non-public definitions.
|
|
|
|
*
|
|
|
|
* 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 IBM Corporation, 2008
|
|
|
|
*
|
|
|
|
* Author: Ingo Molnar <mingo@elte.hu>
|
|
|
|
* Paul E. McKenney <paulmck@linux.vnet.ibm.com>
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/cache.h>
|
|
|
|
#include <linux/spinlock.h>
|
|
|
|
#include <linux/threads.h>
|
|
|
|
#include <linux/cpumask.h>
|
|
|
|
#include <linux/seqlock.h>
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Define shape of hierarchy based on NR_CPUS and CONFIG_RCU_FANOUT.
|
|
|
|
* In theory, it should be possible to add more levels straightforwardly.
|
|
|
|
* In practice, this has not been tested, so there is probably some
|
|
|
|
* bug somewhere.
|
|
|
|
*/
|
|
|
|
#define MAX_RCU_LVLS 3
|
|
|
|
#define RCU_FANOUT (CONFIG_RCU_FANOUT)
|
|
|
|
#define RCU_FANOUT_SQ (RCU_FANOUT * RCU_FANOUT)
|
|
|
|
#define RCU_FANOUT_CUBE (RCU_FANOUT_SQ * RCU_FANOUT)
|
|
|
|
|
|
|
|
#if NR_CPUS <= RCU_FANOUT
|
|
|
|
# define NUM_RCU_LVLS 1
|
|
|
|
# define NUM_RCU_LVL_0 1
|
|
|
|
# define NUM_RCU_LVL_1 (NR_CPUS)
|
|
|
|
# define NUM_RCU_LVL_2 0
|
|
|
|
# define NUM_RCU_LVL_3 0
|
|
|
|
#elif NR_CPUS <= RCU_FANOUT_SQ
|
|
|
|
# define NUM_RCU_LVLS 2
|
|
|
|
# define NUM_RCU_LVL_0 1
|
|
|
|
# define NUM_RCU_LVL_1 (((NR_CPUS) + RCU_FANOUT - 1) / RCU_FANOUT)
|
|
|
|
# define NUM_RCU_LVL_2 (NR_CPUS)
|
|
|
|
# define NUM_RCU_LVL_3 0
|
|
|
|
#elif NR_CPUS <= RCU_FANOUT_CUBE
|
|
|
|
# define NUM_RCU_LVLS 3
|
|
|
|
# define NUM_RCU_LVL_0 1
|
|
|
|
# define NUM_RCU_LVL_1 (((NR_CPUS) + RCU_FANOUT_SQ - 1) / RCU_FANOUT_SQ)
|
|
|
|
# define NUM_RCU_LVL_2 (((NR_CPUS) + (RCU_FANOUT) - 1) / (RCU_FANOUT))
|
|
|
|
# define NUM_RCU_LVL_3 NR_CPUS
|
|
|
|
#else
|
|
|
|
# error "CONFIG_RCU_FANOUT insufficient for NR_CPUS"
|
|
|
|
#endif /* #if (NR_CPUS) <= RCU_FANOUT */
|
|
|
|
|
|
|
|
#define RCU_SUM (NUM_RCU_LVL_0 + NUM_RCU_LVL_1 + NUM_RCU_LVL_2 + NUM_RCU_LVL_3)
|
|
|
|
#define NUM_RCU_NODES (RCU_SUM - NR_CPUS)
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Dynticks per-CPU state.
|
|
|
|
*/
|
|
|
|
struct rcu_dynticks {
|
|
|
|
int dynticks_nesting; /* Track nesting level, sort of. */
|
|
|
|
int dynticks; /* Even value for dynticks-idle, else odd. */
|
|
|
|
int dynticks_nmi; /* Even value for either dynticks-idle or */
|
|
|
|
/* not in nmi handler, else odd. So this */
|
|
|
|
/* remains even for nmi from irq handler. */
|
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Definition for node within the RCU grace-period-detection hierarchy.
|
|
|
|
*/
|
|
|
|
struct rcu_node {
|
|
|
|
spinlock_t lock;
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 16:56:52 -04:00
|
|
|
long gpnum; /* Current grace period for this node. */
|
2009-08-27 18:00:12 -04:00
|
|
|
/* This will either be equal to or one */
|
|
|
|
/* behind the root rcu_node's gpnum. */
|
2009-08-22 16:56:45 -04:00
|
|
|
unsigned long qsmask; /* CPUs or groups that need to switch in */
|
|
|
|
/* order for current grace period to proceed.*/
|
|
|
|
unsigned long qsmaskinit;
|
|
|
|
/* Per-GP initialization for qsmask. */
|
|
|
|
unsigned long grpmask; /* Mask to apply to parent qsmask. */
|
|
|
|
int grplo; /* lowest-numbered CPU or group here. */
|
|
|
|
int grphi; /* highest-numbered CPU or group here. */
|
|
|
|
u8 grpnum; /* CPU/group number for next level up. */
|
|
|
|
u8 level; /* root is at level 0. */
|
|
|
|
struct rcu_node *parent;
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 16:56:52 -04:00
|
|
|
struct list_head blocked_tasks[2];
|
|
|
|
/* Tasks blocked in RCU read-side critsect. */
|
2009-08-22 16:56:45 -04:00
|
|
|
} ____cacheline_internodealigned_in_smp;
|
|
|
|
|
|
|
|
/* Index values for nxttail array in struct rcu_data. */
|
|
|
|
#define RCU_DONE_TAIL 0 /* Also RCU_WAIT head. */
|
|
|
|
#define RCU_WAIT_TAIL 1 /* Also RCU_NEXT_READY head. */
|
|
|
|
#define RCU_NEXT_READY_TAIL 2 /* Also RCU_NEXT head. */
|
|
|
|
#define RCU_NEXT_TAIL 3
|
|
|
|
#define RCU_NEXT_SIZE 4
|
|
|
|
|
|
|
|
/* Per-CPU data for read-copy update. */
|
|
|
|
struct rcu_data {
|
|
|
|
/* 1) quiescent-state and grace-period handling : */
|
|
|
|
long completed; /* Track rsp->completed gp number */
|
|
|
|
/* in order to detect GP end. */
|
|
|
|
long gpnum; /* Highest gp number that this CPU */
|
|
|
|
/* is aware of having started. */
|
|
|
|
long passed_quiesc_completed;
|
|
|
|
/* Value of completed at time of qs. */
|
|
|
|
bool passed_quiesc; /* User-mode/idle loop etc. */
|
|
|
|
bool qs_pending; /* Core waits for quiesc state. */
|
|
|
|
bool beenonline; /* CPU online at least once. */
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 16:56:52 -04:00
|
|
|
bool preemptable; /* Preemptable RCU? */
|
2009-08-22 16:56:45 -04:00
|
|
|
struct rcu_node *mynode; /* This CPU's leaf of hierarchy */
|
|
|
|
unsigned long grpmask; /* Mask to apply to leaf qsmask. */
|
|
|
|
|
|
|
|
/* 2) batch handling */
|
|
|
|
/*
|
|
|
|
* If nxtlist is not NULL, it is partitioned as follows.
|
|
|
|
* Any of the partitions might be empty, in which case the
|
|
|
|
* pointer to that partition will be equal to the pointer for
|
|
|
|
* the following partition. When the list is empty, all of
|
|
|
|
* the nxttail elements point to nxtlist, which is NULL.
|
|
|
|
*
|
|
|
|
* [*nxttail[RCU_NEXT_READY_TAIL], NULL = *nxttail[RCU_NEXT_TAIL]):
|
|
|
|
* Entries that might have arrived after current GP ended
|
|
|
|
* [*nxttail[RCU_WAIT_TAIL], *nxttail[RCU_NEXT_READY_TAIL]):
|
|
|
|
* Entries known to have arrived before current GP ended
|
|
|
|
* [*nxttail[RCU_DONE_TAIL], *nxttail[RCU_WAIT_TAIL]):
|
|
|
|
* Entries that batch # <= ->completed - 1: waiting for current GP
|
|
|
|
* [nxtlist, *nxttail[RCU_DONE_TAIL]):
|
|
|
|
* Entries that batch # <= ->completed
|
|
|
|
* The grace period for these entries has completed, and
|
|
|
|
* the other grace-period-completed entries may be moved
|
|
|
|
* here temporarily in rcu_process_callbacks().
|
|
|
|
*/
|
|
|
|
struct rcu_head *nxtlist;
|
|
|
|
struct rcu_head **nxttail[RCU_NEXT_SIZE];
|
|
|
|
long qlen; /* # of queued callbacks */
|
|
|
|
long blimit; /* Upper limit on a processed batch */
|
|
|
|
|
|
|
|
#ifdef CONFIG_NO_HZ
|
|
|
|
/* 3) dynticks interface. */
|
|
|
|
struct rcu_dynticks *dynticks; /* Shared per-CPU dynticks state. */
|
|
|
|
int dynticks_snap; /* Per-GP tracking for dynticks. */
|
|
|
|
int dynticks_nmi_snap; /* Per-GP tracking for dynticks_nmi. */
|
|
|
|
#endif /* #ifdef CONFIG_NO_HZ */
|
|
|
|
|
|
|
|
/* 4) reasons this CPU needed to be kicked by force_quiescent_state */
|
|
|
|
#ifdef CONFIG_NO_HZ
|
|
|
|
unsigned long dynticks_fqs; /* Kicked due to dynticks idle. */
|
|
|
|
#endif /* #ifdef CONFIG_NO_HZ */
|
|
|
|
unsigned long offline_fqs; /* Kicked due to being offline. */
|
|
|
|
unsigned long resched_ipi; /* Sent a resched IPI. */
|
|
|
|
|
|
|
|
/* 5) __rcu_pending() statistics. */
|
|
|
|
long n_rcu_pending; /* rcu_pending() calls since boot. */
|
|
|
|
long n_rp_qs_pending;
|
|
|
|
long n_rp_cb_ready;
|
|
|
|
long n_rp_cpu_needs_gp;
|
|
|
|
long n_rp_gp_completed;
|
|
|
|
long n_rp_gp_started;
|
|
|
|
long n_rp_need_fqs;
|
|
|
|
long n_rp_need_nothing;
|
|
|
|
|
|
|
|
int cpu;
|
|
|
|
};
|
|
|
|
|
|
|
|
/* Values for signaled field in struct rcu_state. */
|
|
|
|
#define RCU_GP_INIT 0 /* Grace period being initialized. */
|
|
|
|
#define RCU_SAVE_DYNTICK 1 /* Need to scan dyntick state. */
|
|
|
|
#define RCU_FORCE_QS 2 /* Need to force quiescent state. */
|
|
|
|
#ifdef CONFIG_NO_HZ
|
|
|
|
#define RCU_SIGNAL_INIT RCU_SAVE_DYNTICK
|
|
|
|
#else /* #ifdef CONFIG_NO_HZ */
|
|
|
|
#define RCU_SIGNAL_INIT RCU_FORCE_QS
|
|
|
|
#endif /* #else #ifdef CONFIG_NO_HZ */
|
|
|
|
|
|
|
|
#define RCU_JIFFIES_TILL_FORCE_QS 3 /* for rsp->jiffies_force_qs */
|
|
|
|
#ifdef CONFIG_RCU_CPU_STALL_DETECTOR
|
|
|
|
#define RCU_SECONDS_TILL_STALL_CHECK (10 * HZ) /* for rsp->jiffies_stall */
|
|
|
|
#define RCU_SECONDS_TILL_STALL_RECHECK (30 * HZ) /* for rsp->jiffies_stall */
|
|
|
|
#define RCU_STALL_RAT_DELAY 2 /* Allow other CPUs time */
|
|
|
|
/* to take at least one */
|
|
|
|
/* scheduling clock irq */
|
|
|
|
/* before ratting on them. */
|
|
|
|
|
|
|
|
#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* RCU global state, including node hierarchy. This hierarchy is
|
|
|
|
* represented in "heap" form in a dense array. The root (first level)
|
|
|
|
* of the hierarchy is in ->node[0] (referenced by ->level[0]), the second
|
|
|
|
* level in ->node[1] through ->node[m] (->node[1] referenced by ->level[1]),
|
|
|
|
* and the third level in ->node[m+1] and following (->node[m+1] referenced
|
|
|
|
* by ->level[2]). The number of levels is determined by the number of
|
|
|
|
* CPUs and by CONFIG_RCU_FANOUT. Small systems will have a "hierarchy"
|
|
|
|
* consisting of a single rcu_node.
|
|
|
|
*/
|
|
|
|
struct rcu_state {
|
|
|
|
struct rcu_node node[NUM_RCU_NODES]; /* Hierarchy. */
|
|
|
|
struct rcu_node *level[NUM_RCU_LVLS]; /* Hierarchy levels. */
|
|
|
|
u32 levelcnt[MAX_RCU_LVLS + 1]; /* # nodes in each level. */
|
|
|
|
u8 levelspread[NUM_RCU_LVLS]; /* kids/node in each level. */
|
|
|
|
struct rcu_data *rda[NR_CPUS]; /* array of rdp pointers. */
|
|
|
|
|
|
|
|
/* The following fields are guarded by the root rcu_node's lock. */
|
|
|
|
|
|
|
|
u8 signaled ____cacheline_internodealigned_in_smp;
|
|
|
|
/* Force QS state. */
|
|
|
|
long gpnum; /* Current gp number. */
|
|
|
|
long completed; /* # of last completed gp. */
|
|
|
|
spinlock_t onofflock; /* exclude on/offline and */
|
|
|
|
/* starting new GP. */
|
|
|
|
spinlock_t fqslock; /* Only one task forcing */
|
|
|
|
/* quiescent states. */
|
|
|
|
unsigned long jiffies_force_qs; /* Time at which to invoke */
|
|
|
|
/* force_quiescent_state(). */
|
|
|
|
unsigned long n_force_qs; /* Number of calls to */
|
|
|
|
/* force_quiescent_state(). */
|
|
|
|
unsigned long n_force_qs_lh; /* ~Number of calls leaving */
|
|
|
|
/* due to lock unavailable. */
|
|
|
|
unsigned long n_force_qs_ngp; /* Number of calls leaving */
|
|
|
|
/* due to no GP active. */
|
|
|
|
#ifdef CONFIG_RCU_CPU_STALL_DETECTOR
|
|
|
|
unsigned long gp_start; /* Time at which GP started, */
|
|
|
|
/* but in jiffies. */
|
|
|
|
unsigned long jiffies_stall; /* Time at which to check */
|
|
|
|
/* for CPU stalls. */
|
|
|
|
#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */
|
|
|
|
#ifdef CONFIG_NO_HZ
|
|
|
|
long dynticks_completed; /* Value of completed @ snap. */
|
|
|
|
#endif /* #ifdef CONFIG_NO_HZ */
|
|
|
|
};
|
|
|
|
|
|
|
|
#ifdef RCU_TREE_NONCORE
|
2009-03-25 11:42:24 -04:00
|
|
|
|
|
|
|
/*
|
|
|
|
* RCU implementation internal declarations:
|
|
|
|
*/
|
2009-08-22 16:56:46 -04:00
|
|
|
extern struct rcu_state rcu_sched_state;
|
|
|
|
DECLARE_PER_CPU(struct rcu_data, rcu_sched_data);
|
2009-03-25 11:42:24 -04:00
|
|
|
|
|
|
|
extern struct rcu_state rcu_bh_state;
|
|
|
|
DECLARE_PER_CPU(struct rcu_data, rcu_bh_data);
|
|
|
|
|
rcu: Merge preemptable-RCU functionality into hierarchical RCU
Create a kernel/rcutree_plugin.h file that contains definitions
for preemptable RCU (or, under the #else branch of the #ifdef,
empty definitions for the classic non-preemptable semantics).
These definitions fit into plugins defined in kernel/rcutree.c
for this purpose.
This variant of preemptable RCU uses a new algorithm whose
read-side expense is roughly that of classic hierarchical RCU
under CONFIG_PREEMPT. This new algorithm's update-side expense
is similar to that of classic hierarchical RCU, and, in absence
of read-side preemption or blocking, is exactly that of classic
hierarchical RCU. Perhaps more important, this new algorithm
has a much simpler implementation, saving well over 1,000 lines
of code compared to mainline's implementation of preemptable
RCU, which will hopefully be retired in favor of this new
algorithm.
The simplifications are obtained by maintaining per-task
nesting state for running tasks, and using a simple
lock-protected algorithm to handle accounting when tasks block
within RCU read-side critical sections, making use of lessons
learned while creating numerous user-level RCU implementations
over the past 18 months.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: laijs@cn.fujitsu.com
Cc: dipankar@in.ibm.com
Cc: akpm@linux-foundation.org
Cc: mathieu.desnoyers@polymtl.ca
Cc: josht@linux.vnet.ibm.com
Cc: dvhltc@us.ibm.com
Cc: niv@us.ibm.com
Cc: peterz@infradead.org
Cc: rostedt@goodmis.org
LKML-Reference: <12509746134003-git-send-email->
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-22 16:56:52 -04:00
|
|
|
#ifdef CONFIG_TREE_PREEMPT_RCU
|
|
|
|
extern struct rcu_state rcu_preempt_state;
|
|
|
|
DECLARE_PER_CPU(struct rcu_data, rcu_preempt_data);
|
|
|
|
#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */
|
|
|
|
|
2009-08-22 16:56:45 -04:00
|
|
|
#endif /* #ifdef RCU_TREE_NONCORE */
|
|
|
|
|