leandrof/linux
1
0
Fork 0
forked from Minki/linux
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge branch 'rcu/doc' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu into core/urgent

Browse Source 
Pull RCU documentation update for reducing OS jitter due to
per-CPU kthreads, from Paul McKenney.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
This commit is contained in:
Ingo Molnar 2013-04-30 10:49:04 +02:00
commit fd29f424d4
16 changed files with 852 additions and 528 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
Documentation/RCU/checklist.txt
View File

@ -217,9 +217,14 @@ over a rather long period of time, but improvements are always welcome!
whether the increased speed is worth it.
8. Although synchronize_rcu() is slower than is call_rcu(), it
usually results in simpler code. So, unless update performance
is critically important or the updaters cannot block,
synchronize_rcu() should be used in preference to call_rcu().
usually results in simpler code. So, unless update performance is
critically important, the updaters cannot block, or the latency of
synchronize_rcu() is visible from userspace, synchronize_rcu()
should be used in preference to call_rcu(). Furthermore,
kfree_rcu() usually results in even simpler code than does
synchronize_rcu() without synchronize_rcu()'s multi-millisecond
latency. So please take advantage of kfree_rcu()'s "fire and
forget" memory-freeing capabilities where it applies.
An especially important property of the synchronize_rcu()
primitive is that it automatically self-limits: if grace periods
@ -268,7 +273,8 @@ over a rather long period of time, but improvements are always welcome!
e. Periodically invoke synchronize_rcu(), permitting a limited
number of updates per grace period.
The same cautions apply to call_rcu_bh() and call_rcu_sched().
The same cautions apply to call_rcu_bh(), call_rcu_sched(),
call_srcu(), and kfree_rcu().
9. All RCU list-traversal primitives, which include
rcu_dereference(), list_for_each_entry_rcu(), and
@ -296,9 +302,9 @@ over a rather long period of time, but improvements are always welcome!
all currently executing rcu_read_lock()-protected RCU read-side
critical sections complete. It does -not- necessarily guarantee
that all currently running interrupts, NMIs, preempt_disable()
code, or idle loops will complete. Therefore, if you do not have
rcu_read_lock()-protected read-side critical sections, do -not-
use synchronize_rcu().
code, or idle loops will complete. Therefore, if your
read-side critical sections are protected by something other
than rcu_read_lock(), do -not- use synchronize_rcu().
Similarly, disabling preemption is not an acceptable substitute
for rcu_read_lock(). Code that attempts to use preemption
@ -401,9 +407,9 @@ over a rather long period of time, but improvements are always welcome!
read-side critical sections. It is the responsibility of the
RCU update-side primitives to deal with this.
17. Use CONFIG_PROVE_RCU, CONFIG_DEBUG_OBJECTS_RCU_HEAD, and
the __rcu sparse checks to validate your RCU code. These
can help find problems as follows:
17. Use CONFIG_PROVE_RCU, CONFIG_DEBUG_OBJECTS_RCU_HEAD, and the
__rcu sparse checks (enabled by CONFIG_SPARSE_RCU_POINTER) to
validate your RCU code. These can help find problems as follows:
CONFIG_PROVE_RCU: check that accesses to RCU-protected data
structures are carried out under the proper RCU

5
Documentation/RCU/lockdep.txt
View File

@ -64,6 +64,11 @@ checking of rcu_dereference() primitives:
but retain the compiler constraints that prevent duplicating
or coalescsing. This is useful when when testing the
value of the pointer itself, for example, against NULL.
rcu_access_index(idx):
Return the value of the index and omit all barriers, but
retain the compiler constraints that prevent duplicating
or coalescsing. This is useful when when testing the
value of the index itself, for example, against -1.
The rcu_dereference_check() check expression can be any boolean
expression, but would normally include a lockdep expression. However,

15
Documentation/RCU/rcubarrier.txt
View File

@ -79,7 +79,20 @@ complete. Pseudo-code using rcu_barrier() is as follows:
2. Execute rcu_barrier().
3. Allow the module to be unloaded.
The rcutorture module makes use of rcu_barrier in its exit function
There are also rcu_barrier_bh(), rcu_barrier_sched(), and srcu_barrier()
functions for the other flavors of RCU, and you of course must match
the flavor of rcu_barrier() with that of call_rcu(). If your module
uses multiple flavors of call_rcu(), then it must also use multiple
flavors of rcu_barrier() when unloading that module. For example, if
it uses call_rcu_bh(), call_srcu() on srcu_struct_1, and call_srcu() on
srcu_struct_2(), then the following three lines of code will be required
when unloading:
1 rcu_barrier_bh();
2 srcu_barrier(&srcu_struct_1);
3 srcu_barrier(&srcu_struct_2);
The rcutorture module makes use of rcu_barrier() in its exit function
as follows:
1 static void

33
Documentation/RCU/stallwarn.txt
View File

@ -92,14 +92,14 @@ If the CONFIG_RCU_CPU_STALL_INFO kernel configuration parameter is set,
more information is printed with the stall-warning message, for example:
INFO: rcu_preempt detected stall on CPU
0: (63959 ticks this GP) idle=241/3fffffffffffffff/0
0: (63959 ticks this GP) idle=241/3fffffffffffffff/0 softirq=82/543
(t=65000 jiffies)
In kernels with CONFIG_RCU_FAST_NO_HZ, even more information is
printed:
INFO: rcu_preempt detected stall on CPU
0: (64628 ticks this GP) idle=dd5/3fffffffffffffff/0 drain=0 . timer not pending
0: (64628 ticks this GP) idle=dd5/3fffffffffffffff/0 softirq=82/543 last_accelerate: a345/d342 nonlazy_posted: 25 .D
(t=65000 jiffies)
The "(64628 ticks this GP)" indicates that this CPU has taken more
@ -116,13 +116,28 @@ number between the two "/"s is the value of the nesting, which will
be a small positive number if in the idle loop and a very large positive
number (as shown above) otherwise.
For CONFIG_RCU_FAST_NO_HZ kernels, the "drain=0" indicates that the CPU is
not in the process of trying to force itself into dyntick-idle state, the
"." indicates that the CPU has not given up forcing RCU into dyntick-idle
mode (it would be "H" otherwise), and the "timer not pending" indicates
that the CPU has not recently forced RCU into dyntick-idle mode (it
would otherwise indicate the number of microseconds remaining in this
forced state).
The "softirq=" portion of the message tracks the number of RCU softirq
handlers that the stalled CPU has executed. The number before the "/"
is the number that had executed since boot at the time that this CPU
last noted the beginning of a grace period, which might be the current
(stalled) grace period, or it might be some earlier grace period (for
example, if the CPU might have been in dyntick-idle mode for an extended
time period. The number after the "/" is the number that have executed
since boot until the current time. If this latter number stays constant
across repeated stall-warning messages, it is possible that RCU's softirq
handlers are no longer able to execute on this CPU. This can happen if
the stalled CPU is spinning with interrupts are disabled, or, in -rt
kernels, if a high-priority process is starving RCU's softirq handler.
For CONFIG_RCU_FAST_NO_HZ kernels, the "last_accelerate:" prints the
low-order 16 bits (in hex) of the jiffies counter when this CPU last
invoked rcu_try_advance_all_cbs() from rcu_needs_cpu() or last invoked
rcu_accelerate_cbs() from rcu_prepare_for_idle(). The "nonlazy_posted:"
prints the number of non-lazy callbacks posted since the last call to
rcu_needs_cpu(). Finally, an "L" indicates that there are currently
no non-lazy callbacks ("." is printed otherwise, as shown above) and
"D" indicates that dyntick-idle processing is enabled ("." is printed
otherwise, for example, if disabled via the "nohz=" kernel boot parameter).
Multiple Warnings From One Stall

4
Documentation/RCU/whatisRCU.txt
View File

@ -265,9 +265,9 @@ rcu_dereference()
rcu_read_lock();
p = rcu_dereference(head.next);
rcu_read_unlock();
x = p->address;
x = p->address; /* BUG!!! */
rcu_read_lock();
y = p->data;
y = p->data; /* BUG!!! */
rcu_read_unlock();
Holding a reference from one RCU read-side critical section

35
Documentation/kernel-parameters.txt
View File

@ -2484,9 +2484,12 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
In kernels built with CONFIG_RCU_NOCB_CPU=y, set
the specified list of CPUs to be no-callback CPUs.
Invocation of these CPUs' RCU callbacks will
be offloaded to "rcuoN" kthreads created for
that purpose. This reduces OS jitter on the
be offloaded to "rcuox/N" kthreads created for
that purpose, where "x" is "b" for RCU-bh, "p"
for RCU-preempt, and "s" for RCU-sched, and "N"
is the CPU number. This reduces OS jitter on the
offloaded CPUs, which can be useful for HPC and
real-time workloads. It can also improve energy
efficiency for asymmetric multiprocessors.
@ -2510,6 +2513,17 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
leaf rcu_node structure. Useful for very large
systems.
rcutree.jiffies_till_first_fqs= [KNL,BOOT]
Set delay from grace-period initialization to
first attempt to force quiescent states.
Units are jiffies, minimum value is zero,
and maximum value is HZ.
rcutree.jiffies_till_next_fqs= [KNL,BOOT]
Set delay between subsequent attempts to force
quiescent states. Units are jiffies, minimum
value is one, and maximum value is HZ.
rcutree.qhimark= [KNL,BOOT]
Set threshold of queued
RCU callbacks over which batch limiting is disabled.
@ -2524,16 +2538,15 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
rcutree.rcu_cpu_stall_timeout= [KNL,BOOT]
Set timeout for RCU CPU stall warning messages.
rcutree.jiffies_till_first_fqs= [KNL,BOOT]
Set delay from grace-period initialization to
first attempt to force quiescent states.
Units are jiffies, minimum value is zero,
and maximum value is HZ.
rcutree.rcu_idle_gp_delay= [KNL,BOOT]
Set wakeup interval for idle CPUs that have
RCU callbacks (RCU_FAST_NO_HZ=y).
rcutree.jiffies_till_next_fqs= [KNL,BOOT]
Set delay between subsequent attempts to force
quiescent states. Units are jiffies, minimum
value is one, and maximum value is HZ.
rcutree.rcu_idle_lazy_gp_delay= [KNL,BOOT]
Set wakeup interval for idle CPUs that have
only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y).
Lazy RCU callbacks are those which RCU can
prove do nothing more than free memory.
rcutorture.fqs_duration= [KNL,BOOT]
Set duration of force_quiescent_state bursts.

202
Documentation/kernel-per-CPU-kthreads.txt Normal file
View File

@ -0,0 +1,202 @@
REDUCING OS JITTER DUE TO PER-CPU KTHREADS
This document lists per-CPU kthreads in the Linux kernel and presents
options to control their OS jitter. Note that non-per-CPU kthreads are
not listed here. To reduce OS jitter from non-per-CPU kthreads, bind
them to a "housekeeping" CPU dedicated to such work.
REFERENCES
o Documentation/IRQ-affinity.txt: Binding interrupts to sets of CPUs.
o Documentation/cgroups: Using cgroups to bind tasks to sets of CPUs.
o man taskset: Using the taskset command to bind tasks to sets
of CPUs.
o man sched_setaffinity: Using the sched_setaffinity() system
call to bind tasks to sets of CPUs.
o /sys/devices/system/cpu/cpuN/online: Control CPU N's hotplug state,
writing "0" to offline and "1" to online.
o In order to locate kernel-generated OS jitter on CPU N:
cd /sys/kernel/debug/tracing
echo 1 > max_graph_depth # Increase the "1" for more detail
echo function_graph > current_tracer
# run workload
cat per_cpu/cpuN/trace
KTHREADS
Name: ehca_comp/%u
Purpose: Periodically process Infiniband-related work.
To reduce its OS jitter, do any of the following:
1. Don't use eHCA Infiniband hardware, instead choosing hardware
that does not require per-CPU kthreads. This will prevent these
kthreads from being created in the first place. (This will
work for most people, as this hardware, though important, is
relatively old and is produced in relatively low unit volumes.)
2. Do all eHCA-Infiniband-related work on other CPUs, including
interrupts.
3. Rework the eHCA driver so that its per-CPU kthreads are
provisioned only on selected CPUs.
Name: irq/%d-%s
Purpose: Handle threaded interrupts.
To reduce its OS jitter, do the following:
1. Use irq affinity to force the irq threads to execute on
some other CPU.
Name: kcmtpd_ctr_%d
Purpose: Handle Bluetooth work.
To reduce its OS jitter, do one of the following:
1. Don't use Bluetooth, in which case these kthreads won't be
created in the first place.
2. Use irq affinity to force Bluetooth-related interrupts to
occur on some other CPU and furthermore initiate all
Bluetooth activity on some other CPU.
Name: ksoftirqd/%u
Purpose: Execute softirq handlers when threaded or when under heavy load.
To reduce its OS jitter, each softirq vector must be handled
separately as follows:
TIMER_SOFTIRQ: Do all of the following:
1. To the extent possible, keep the CPU out of the kernel when it
is non-idle, for example, by avoiding system calls and by forcing
both kernel threads and interrupts to execute elsewhere.
2. Build with CONFIG_HOTPLUG_CPU=y. After boot completes, force
the CPU offline, then bring it back online. This forces
recurring timers to migrate elsewhere. If you are concerned
with multiple CPUs, force them all offline before bringing the
first one back online. Once you have onlined the CPUs in question,
do not offline any other CPUs, because doing so could force the
timer back onto one of the CPUs in question.
NET_TX_SOFTIRQ and NET_RX_SOFTIRQ: Do all of the following:
1. Force networking interrupts onto other CPUs.
2. Initiate any network I/O on other CPUs.
3. Once your application has started, prevent CPU-hotplug operations
from being initiated from tasks that might run on the CPU to
be de-jittered. (It is OK to force this CPU offline and then
bring it back online before you start your application.)
BLOCK_SOFTIRQ: Do all of the following:
1. Force block-device interrupts onto some other CPU.
2. Initiate any block I/O on other CPUs.
3. Once your application has started, prevent CPU-hotplug operations
from being initiated from tasks that might run on the CPU to
be de-jittered. (It is OK to force this CPU offline and then
bring it back online before you start your application.)
BLOCK_IOPOLL_SOFTIRQ: Do all of the following:
1. Force block-device interrupts onto some other CPU.
2. Initiate any block I/O and block-I/O polling on other CPUs.
3. Once your application has started, prevent CPU-hotplug operations
from being initiated from tasks that might run on the CPU to
be de-jittered. (It is OK to force this CPU offline and then
bring it back online before you start your application.)
TASKLET_SOFTIRQ: Do one or more of the following:
1. Avoid use of drivers that use tasklets. (Such drivers will contain
calls to things like tasklet_schedule().)
2. Convert all drivers that you must use from tasklets to workqueues.
3. Force interrupts for drivers using tasklets onto other CPUs,
and also do I/O involving these drivers on other CPUs.
SCHED_SOFTIRQ: Do all of the following:
1. Avoid sending scheduler IPIs to the CPU to be de-jittered,
for example, ensure that at most one runnable kthread is present
on that CPU. If a thread that expects to run on the de-jittered
CPU awakens, the scheduler will send an IPI that can result in
a subsequent SCHED_SOFTIRQ.
2. Build with CONFIG_RCU_NOCB_CPU=y, CONFIG_RCU_NOCB_CPU_ALL=y,
CONFIG_NO_HZ_FULL=y, and, in addition, ensure that the CPU
to be de-jittered is marked as an adaptive-ticks CPU using the
"nohz_full=" boot parameter. This reduces the number of
scheduler-clock interrupts that the de-jittered CPU receives,
minimizing its chances of being selected to do the load balancing
work that runs in SCHED_SOFTIRQ context.
3. To the extent possible, keep the CPU out of the kernel when it
is non-idle, for example, by avoiding system calls and by
forcing both kernel threads and interrupts to execute elsewhere.
This further reduces the number of scheduler-clock interrupts
received by the de-jittered CPU.
HRTIMER_SOFTIRQ: Do all of the following:
1. To the extent possible, keep the CPU out of the kernel when it
is non-idle. For example, avoid system calls and force both
kernel threads and interrupts to execute elsewhere.
2. Build with CONFIG_HOTPLUG_CPU=y. Once boot completes, force the
CPU offline, then bring it back online. This forces recurring
timers to migrate elsewhere. If you are concerned with multiple
CPUs, force them all offline before bringing the first one
back online. Once you have onlined the CPUs in question, do not
offline any other CPUs, because doing so could force the timer
back onto one of the CPUs in question.
RCU_SOFTIRQ: Do at least one of the following:
1. Offload callbacks and keep the CPU in either dyntick-idle or
adaptive-ticks state by doing all of the following:
a. Build with CONFIG_RCU_NOCB_CPU=y, CONFIG_RCU_NOCB_CPU_ALL=y,
CONFIG_NO_HZ_FULL=y, and, in addition ensure that the CPU
to be de-jittered is marked as an adaptive-ticks CPU using
the "nohz_full=" boot parameter. Bind the rcuo kthreads
to housekeeping CPUs, which can tolerate OS jitter.
b. To the extent possible, keep the CPU out of the kernel
when it is non-idle, for example, by avoiding system
calls and by forcing both kernel threads and interrupts
to execute elsewhere.
2. Enable RCU to do its processing remotely via dyntick-idle by
doing all of the following:
a. Build with CONFIG_NO_HZ=y and CONFIG_RCU_FAST_NO_HZ=y.
b. Ensure that the CPU goes idle frequently, allowing other
CPUs to detect that it has passed through an RCU quiescent
state. If the kernel is built with CONFIG_NO_HZ_FULL=y,
userspace execution also allows other CPUs to detect that
the CPU in question has passed through a quiescent state.
c. To the extent possible, keep the CPU out of the kernel
when it is non-idle, for example, by avoiding system
calls and by forcing both kernel threads and interrupts
to execute elsewhere.
Name: rcuc/%u
Purpose: Execute RCU callbacks in CONFIG_RCU_BOOST=y kernels.
To reduce its OS jitter, do at least one of the following:
1. Build the kernel with CONFIG_PREEMPT=n. This prevents these
kthreads from being created in the first place, and also obviates
the need for RCU priority boosting. This approach is feasible
for workloads that do not require high degrees of responsiveness.
2. Build the kernel with CONFIG_RCU_BOOST=n. This prevents these
kthreads from being created in the first place. This approach
is feasible only if your workload never requires RCU priority
boosting, for example, if you ensure frequent idle time on all
CPUs that might execute within the kernel.
3. Build with CONFIG_RCU_NOCB_CPU=y and CONFIG_RCU_NOCB_CPU_ALL=y,
which offloads all RCU callbacks to kthreads that can be moved
off of CPUs susceptible to OS jitter. This approach prevents the
rcuc/%u kthreads from having any work to do, so that they are
never awakened.
4. Ensure that the CPU never enters the kernel, and, in particular,
avoid initiating any CPU hotplug operations on this CPU. This is
another way of preventing any callbacks from being queued on the
CPU, again preventing the rcuc/%u kthreads from having any work
to do.
Name: rcuob/%d, rcuop/%d, and rcuos/%d
Purpose: Offload RCU callbacks from the corresponding CPU.
To reduce its OS jitter, do at least one of the following:
1. Use affinity, cgroups, or other mechanism to force these kthreads
to execute on some other CPU.
2. Build with CONFIG_RCU_NOCB_CPUS=n, which will prevent these
kthreads from being created in the first place. However, please
note that this will not eliminate OS jitter, but will instead
shift it to RCU_SOFTIRQ.
Name: watchdog/%u
Purpose: Detect software lockups on each CPU.
To reduce its OS jitter, do at least one of the following:
1. Build with CONFIG_LOCKUP_DETECTOR=n, which will prevent these
kthreads from being created in the first place.
2. Echo a zero to /proc/sys/kernel/watchdog to disable the
watchdog timer.
3. Echo a large number of /proc/sys/kernel/watchdog_thresh in
order to reduce the frequency of OS jitter due to the watchdog
timer down to a level that is acceptable for your workload.

5
include/linux/list_bl.h
View File

@ -125,6 +125,11 @@ static inline void hlist_bl_unlock(struct hlist_bl_head *b)
__bit_spin_unlock(0, (unsigned long *)b);
}
static inline bool hlist_bl_is_locked(struct hlist_bl_head *b)
{
return bit_spin_is_locked(0, (unsigned long *)b);
}
/**
* hlist_bl_for_each_entry - iterate over list of given type
* @tpos: the type * to use as a loop cursor.

2
include/linux/rculist_bl.h
View File

@ -20,7 +20,7 @@ static inline void hlist_bl_set_first_rcu(struct hlist_bl_head *h,
static inline struct hlist_bl_node *hlist_bl_first_rcu(struct hlist_bl_head *h)
{
return (struct hlist_bl_node *)
((unsigned long)rcu_dereference(h->first) & ~LIST_BL_LOCKMASK);
((unsigned long)rcu_dereference_check(h->first, hlist_bl_is_locked(h)) & ~LIST_BL_LOCKMASK);
}
/**

1
include/linux/rcupdate.h
View File

@ -80,6 +80,7 @@ extern void do_trace_rcu_torture_read(char *rcutorturename,
#define UINT_CMP_LT(a, b) (UINT_MAX / 2 < (a) - (b))
#define ULONG_CMP_GE(a, b) (ULONG_MAX / 2 >= (a) - (b))
#define ULONG_CMP_LT(a, b) (ULONG_MAX / 2 < (a) - (b))
#define ulong2long(a) (*(long *)(&(a)))
/* Exported common interfaces */

55
include/trace/events/rcu.h
View File

@ -71,6 +71,58 @@ TRACE_EVENT(rcu_grace_period,
__entry->rcuname, __entry->gpnum, __entry->gpevent)
);
/*
* Tracepoint for future grace-period events, including those for no-callbacks
* CPUs. The caller should pull the data from the rcu_node structure,
* other than rcuname, which comes from the rcu_state structure, and event,
* which is one of the following:
*
* "Startleaf": Request a nocb grace period based on leaf-node data.
* "Startedleaf": Leaf-node start proved sufficient.
* "Startedleafroot": Leaf-node start proved sufficient after checking root.
* "Startedroot": Requested a nocb grace period based on root-node data.
* "StartWait": Start waiting for the requested grace period.
* "ResumeWait": Resume waiting after signal.
* "EndWait": Complete wait.
* "Cleanup": Clean up rcu_node structure after previous GP.
* "CleanupMore": Clean up, and another no-CB GP is needed.
*/
TRACE_EVENT(rcu_future_grace_period,
TP_PROTO(char *rcuname, unsigned long gpnum, unsigned long completed,
unsigned long c, u8 level, int grplo, int grphi,
char *gpevent),
TP_ARGS(rcuname, gpnum, completed, c, level, grplo, grphi, gpevent),
TP_STRUCT__entry(
__field(char *, rcuname)
__field(unsigned long, gpnum)
__field(unsigned long, completed)
__field(unsigned long, c)
__field(u8, level)
__field(int, grplo)
__field(int, grphi)
__field(char *, gpevent)
),
TP_fast_assign(
__entry->rcuname = rcuname;
__entry->gpnum = gpnum;
__entry->completed = completed;
__entry->c = c;
__entry->level = level;
__entry->grplo = grplo;
__entry->grphi = grphi;
__entry->gpevent = gpevent;
),
TP_printk("%s %lu %lu %lu %u %d %d %s",
__entry->rcuname, __entry->gpnum, __entry->completed,
__entry->c, __entry->level, __entry->grplo, __entry->grphi,
__entry->gpevent)
);
/*
* Tracepoint for grace-period-initialization events. These are
* distinguished by the type of RCU, the new grace-period number, the
@ -601,6 +653,9 @@ TRACE_EVENT(rcu_barrier,
#define trace_rcu_grace_period(rcuname, gpnum, gpevent) do { } while (0)
#define trace_rcu_grace_period_init(rcuname, gpnum, level, grplo, grphi, \
qsmask) do { } while (0)
#define trace_rcu_future_grace_period(rcuname, gpnum, completed, c, \
level, grplo, grphi, event) \
do { } while (0)
#define trace_rcu_preempt_task(rcuname, pid, gpnum) do { } while (0)
#define trace_rcu_unlock_preempted_task(rcuname, gpnum, pid) do { } while (0)
#define trace_rcu_quiescent_state_report(rcuname, gpnum, mask, qsmask, level, \

71
init/Kconfig
View File

@ -578,13 +578,16 @@ config RCU_FAST_NO_HZ
depends on NO_HZ && SMP
default n
help
This option causes RCU to attempt to accelerate grace periods in
order to allow CPUs to enter dynticks-idle state more quickly.
On the other hand, this option increases the overhead of the
dynticks-idle checking, thus degrading scheduling latency.
This option permits CPUs to enter dynticks-idle state even if
they have RCU callbacks queued, and prevents RCU from waking
these CPUs up more than roughly once every four jiffies (by
default, you can adjust this using the rcutree.rcu_idle_gp_delay
parameter), thus improving energy efficiency. On the other
hand, this option increases the duration of RCU grace periods,
for example, slowing down synchronize_rcu().
Say Y if energy efficiency is critically important, and you don't
care about real-time response.
Say Y if energy efficiency is critically important, and you
don't care about increased grace-period durations.
Say N if you are unsure.
@ -651,7 +654,7 @@ config RCU_BOOST_DELAY
Accept the default if unsure.
config RCU_NOCB_CPU
bool "Offload RCU callback processing from boot-selected CPUs"
bool "Offload RCU callback processing from boot-selected CPUs (EXPERIMENTAL"
depends on TREE_RCU || TREE_PREEMPT_RCU
default n
help
@ -662,16 +665,56 @@ config RCU_NOCB_CPU
This option offloads callback invocation from the set of
CPUs specified at boot time by the rcu_nocbs parameter.
For each such CPU, a kthread ("rcuoN") will be created to
invoke callbacks, where the "N" is the CPU being offloaded.
Nothing prevents this kthread from running on the specified
CPUs, but (1) the kthreads may be preempted between each
callback, and (2) affinity or cgroups can be used to force
the kthreads to run on whatever set of CPUs is desired.
For each such CPU, a kthread ("rcuox/N") will be created to
invoke callbacks, where the "N" is the CPU being offloaded,
and where the "x" is "b" for RCU-bh, "p" for RCU-preempt, and
"s" for RCU-sched. Nothing prevents this kthread from running
on the specified CPUs, but (1) the kthreads may be preempted
between each callback, and (2) affinity or cgroups can be used
to force the kthreads to run on whatever set of CPUs is desired.
Say Y here if you want reduced OS jitter on selected CPUs.
Say Y here if you want to help to debug reduced OS jitter.
Say N here if you are unsure.
choice
prompt "Build-forced no-CBs CPUs"
default RCU_NOCB_CPU_NONE
help
This option allows no-CBs CPUs to be specified at build time.
Additional no-CBs CPUs may be specified by the rcu_nocbs=
boot parameter.
config RCU_NOCB_CPU_NONE
bool "No build_forced no-CBs CPUs"
depends on RCU_NOCB_CPU
help
This option does not force any of the CPUs to be no-CBs CPUs.
Only CPUs designated by the rcu_nocbs= boot parameter will be
no-CBs CPUs.
config RCU_NOCB_CPU_ZERO
bool "CPU 0 is a build_forced no-CBs CPU"
depends on RCU_NOCB_CPU
help
This option forces CPU 0 to be a no-CBs CPU. Additional CPUs
may be designated as no-CBs CPUs using the rcu_nocbs= boot
parameter will be no-CBs CPUs.
Select this if CPU 0 needs to be a no-CBs CPU for real-time
or energy-efficiency reasons.
config RCU_NOCB_CPU_ALL
bool "All CPUs are build_forced no-CBs CPUs"
depends on RCU_NOCB_CPU
help
This option forces all CPUs to be no-CBs CPUs. The rcu_nocbs=
boot parameter will be ignored.
Select this if all CPUs need to be no-CBs CPUs for real-time
or energy-efficiency reasons.
endchoice
endmenu # "RCU Subsystem"
config IKCONFIG

260
kernel/rcutree.c
View File

@ -64,7 +64,7 @@
static struct lock_class_key rcu_node_class[RCU_NUM_LVLS];
static struct lock_class_key rcu_fqs_class[RCU_NUM_LVLS];
#define RCU_STATE_INITIALIZER(sname, cr) { \
#define RCU_STATE_INITIALIZER(sname, sabbr, cr) { \
.level = { &sname##_state.node[0] }, \
.call = cr, \
.fqs_state = RCU_GP_IDLE, \
@ -76,13 +76,14 @@ static struct lock_class_key rcu_fqs_class[RCU_NUM_LVLS];
.barrier_mutex = __MUTEX_INITIALIZER(sname##_state.barrier_mutex), \
.onoff_mutex = __MUTEX_INITIALIZER(sname##_state.onoff_mutex), \
.name = #sname, \
.abbr = sabbr, \
}
struct rcu_state rcu_sched_state =
RCU_STATE_INITIALIZER(rcu_sched, call_rcu_sched);
RCU_STATE_INITIALIZER(rcu_sched, 's', call_rcu_sched);
DEFINE_PER_CPU(struct rcu_data, rcu_sched_data);
struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh, call_rcu_bh);
struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh, 'b', call_rcu_bh);
DEFINE_PER_CPU(struct rcu_data, rcu_bh_data);
static struct rcu_state *rcu_state;
@ -223,6 +224,8 @@ static ulong jiffies_till_next_fqs = RCU_JIFFIES_TILL_FORCE_QS;
module_param(jiffies_till_first_fqs, ulong, 0644);
module_param(jiffies_till_next_fqs, ulong, 0644);
static void rcu_start_gp_advanced(struct rcu_state *rsp, struct rcu_node *rnp,
struct rcu_data *rdp);
static void force_qs_rnp(struct rcu_state *rsp, int (*f)(struct rcu_data *));
static void force_quiescent_state(struct rcu_state *rsp);
static int rcu_pending(int cpu);
@ -310,6 +313,8 @@ cpu_needs_another_gp(struct rcu_state *rsp, struct rcu_data *rdp)
if (rcu_gp_in_progress(rsp))
return 0; /* No, a grace period is already in progress. */
if (rcu_nocb_needs_gp(rsp))
return 1; /* Yes, a no-CBs CPU needs one. */
if (!rdp->nxttail[RCU_NEXT_TAIL])
return 0; /* No, this is a no-CBs (or offline) CPU. */
if (*rdp->nxttail[RCU_NEXT_READY_TAIL])
@ -1035,10 +1040,11 @@ static void init_callback_list(struct rcu_data *rdp)
{
int i;
if (init_nocb_callback_list(rdp))
return;
rdp->nxtlist = NULL;
for (i = 0; i < RCU_NEXT_SIZE; i++)
rdp->nxttail[i] = &rdp->nxtlist;
init_nocb_callback_list(rdp);
}
/*
@ -1070,6 +1076,120 @@ static unsigned long rcu_cbs_completed(struct rcu_state *rsp,
return rnp->completed + 2;
}
/*
* Trace-event helper function for rcu_start_future_gp() and
* rcu_nocb_wait_gp().
*/
static void trace_rcu_future_gp(struct rcu_node *rnp, struct rcu_data *rdp,
unsigned long c, char *s)
{
trace_rcu_future_grace_period(rdp->rsp->name, rnp->gpnum,
rnp->completed, c, rnp->level,
rnp->grplo, rnp->grphi, s);
}
/*
* Start some future grace period, as needed to handle newly arrived
* callbacks. The required future grace periods are recorded in each
* rcu_node structure's ->need_future_gp field.
*
* The caller must hold the specified rcu_node structure's ->lock.
*/
static unsigned long __maybe_unused
rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp)
{
unsigned long c;
int i;
struct rcu_node *rnp_root = rcu_get_root(rdp->rsp);
/*
* Pick up grace-period number for new callbacks. If this
* grace period is already marked as needed, return to the caller.
*/
c = rcu_cbs_completed(rdp->rsp, rnp);
trace_rcu_future_gp(rnp, rdp, c, "Startleaf");
if (rnp->need_future_gp[c & 0x1]) {
trace_rcu_future_gp(rnp, rdp, c, "Prestartleaf");
return c;
}
/*
* If either this rcu_node structure or the root rcu_node structure
* believe that a grace period is in progress, then we must wait
* for the one following, which is in "c". Because our request
* will be noticed at the end of the current grace period, we don't
* need to explicitly start one.
*/
if (rnp->gpnum != rnp->completed ||
ACCESS_ONCE(rnp->gpnum) != ACCESS_ONCE(rnp->completed)) {
rnp->need_future_gp[c & 0x1]++;
trace_rcu_future_gp(rnp, rdp, c, "Startedleaf");
return c;
}
/*
* There might be no grace period in progress. If we don't already
* hold it, acquire the root rcu_node structure's lock in order to
* start one (if needed).
*/
if (rnp != rnp_root)
raw_spin_lock(&rnp_root->lock);
/*
* Get a new grace-period number. If there really is no grace
* period in progress, it will be smaller than the one we obtained
* earlier. Adjust callbacks as needed. Note that even no-CBs
* CPUs have a ->nxtcompleted[] array, so no no-CBs checks needed.
*/
c = rcu_cbs_completed(rdp->rsp, rnp_root);
for (i = RCU_DONE_TAIL; i < RCU_NEXT_TAIL; i++)
if (ULONG_CMP_LT(c, rdp->nxtcompleted[i]))
rdp->nxtcompleted[i] = c;
/*
* If the needed for the required grace period is already
* recorded, trace and leave.
*/
if (rnp_root->need_future_gp[c & 0x1]) {
trace_rcu_future_gp(rnp, rdp, c, "Prestartedroot");
goto unlock_out;
}
/* Record the need for the future grace period. */
rnp_root->need_future_gp[c & 0x1]++;
/* If a grace period is not already in progress, start one. */
if (rnp_root->gpnum != rnp_root->completed) {
trace_rcu_future_gp(rnp, rdp, c, "Startedleafroot");
} else {
trace_rcu_future_gp(rnp, rdp, c, "Startedroot");
rcu_start_gp_advanced(rdp->rsp, rnp_root, rdp);
}
unlock_out:
if (rnp != rnp_root)
raw_spin_unlock(&rnp_root->lock);
return c;
}
/*
* Clean up any old requests for the just-ended grace period. Also return
* whether any additional grace periods have been requested. Also invoke
* rcu_nocb_gp_cleanup() in order to wake up any no-callbacks kthreads
* waiting for this grace period to complete.
*/
static int rcu_future_gp_cleanup(struct rcu_state *rsp, struct rcu_node *rnp)
{
int c = rnp->completed;
int needmore;
struct rcu_data *rdp = this_cpu_ptr(rsp->rda);
rcu_nocb_gp_cleanup(rsp, rnp);
rnp->need_future_gp[c & 0x1] = 0;
needmore = rnp->need_future_gp[(c + 1) & 0x1];
trace_rcu_future_gp(rnp, rdp, c, needmore ? "CleanupMore" : "Cleanup");
return needmore;
}
/*
* If there is room, assign a ->completed number to any callbacks on
* this CPU that have not already been assigned. Also accelerate any
@ -1129,6 +1249,8 @@ static void rcu_accelerate_cbs(struct rcu_state *rsp, struct rcu_node *rnp,
rdp->nxttail[i] = rdp->nxttail[RCU_NEXT_TAIL];
rdp->nxtcompleted[i] = c;
}
/* Record any needed additional grace periods. */
rcu_start_future_gp(rnp, rdp);
/* Trace depending on how much we were able to accelerate. */
if (!*rdp->nxttail[RCU_WAIT_TAIL])
@ -1308,9 +1430,9 @@ static int rcu_gp_init(struct rcu_state *rsp)
rdp = this_cpu_ptr(rsp->rda);
rcu_preempt_check_blocked_tasks(rnp);
rnp->qsmask = rnp->qsmaskinit;
rnp->gpnum = rsp->gpnum;
ACCESS_ONCE(rnp->gpnum) = rsp->gpnum;
WARN_ON_ONCE(rnp->completed != rsp->completed);
rnp->completed = rsp->completed;
ACCESS_ONCE(rnp->completed) = rsp->completed;
if (rnp == rdp->mynode)
rcu_start_gp_per_cpu(rsp, rnp, rdp);
rcu_preempt_boost_start_gp(rnp);
@ -1319,7 +1441,8 @@ static int rcu_gp_init(struct rcu_state *rsp)
rnp->grphi, rnp->qsmask);
raw_spin_unlock_irq(&rnp->lock);
#ifdef CONFIG_PROVE_RCU_DELAY
if ((random32() % (rcu_num_nodes * 8)) == 0)
if ((random32() % (rcu_num_nodes * 8)) == 0 &&
system_state == SYSTEM_RUNNING)
schedule_timeout_uninterruptible(2);
#endif /* #ifdef CONFIG_PROVE_RCU_DELAY */
cond_resched();
@ -1361,6 +1484,7 @@ int rcu_gp_fqs(struct rcu_state *rsp, int fqs_state_in)
static void rcu_gp_cleanup(struct rcu_state *rsp)
{
unsigned long gp_duration;
int nocb = 0;
struct rcu_data *rdp;
struct rcu_node *rnp = rcu_get_root(rsp);
@ -1390,17 +1514,23 @@ static void rcu_gp_cleanup(struct rcu_state *rsp)
*/
rcu_for_each_node_breadth_first(rsp, rnp) {
raw_spin_lock_irq(&rnp->lock);
rnp->completed = rsp->gpnum;
ACCESS_ONCE(rnp->completed) = rsp->gpnum;
rdp = this_cpu_ptr(rsp->rda);
if (rnp == rdp->mynode)
__rcu_process_gp_end(rsp, rnp, rdp);
nocb += rcu_future_gp_cleanup(rsp, rnp);
raw_spin_unlock_irq(&rnp->lock);
cond_resched();
}
rnp = rcu_get_root(rsp);
raw_spin_lock_irq(&rnp->lock);
rcu_nocb_gp_set(rnp, nocb);
rsp->completed = rsp->gpnum; /* Declare grace period done. */
trace_rcu_grace_period(rsp->name, rsp->completed, "end");
rsp->fqs_state = RCU_GP_IDLE;
rdp = this_cpu_ptr(rsp->rda);
rcu_advance_cbs(rsp, rnp, rdp); /* Reduce false positives below. */
if (cpu_needs_another_gp(rsp, rdp))
rsp->gp_flags = 1;
raw_spin_unlock_irq(&rnp->lock);
@ -1476,57 +1606,62 @@ static int __noreturn rcu_gp_kthread(void *arg)
/*
* Start a new RCU grace period if warranted, re-initializing the hierarchy
* in preparation for detecting the next grace period. The caller must hold
* the root node's ->lock, which is released before return. Hard irqs must
* be disabled.
* the root node's ->lock and hard irqs must be disabled.
*
* Note that it is legal for a dying CPU (which is marked as offline) to
* invoke this function. This can happen when the dying CPU reports its
* quiescent state.
*/
static void
rcu_start_gp(struct rcu_state *rsp, unsigned long flags)
__releases(rcu_get_root(rsp)->lock)
rcu_start_gp_advanced(struct rcu_state *rsp, struct rcu_node *rnp,
struct rcu_data *rdp)
{
struct rcu_data *rdp = this_cpu_ptr(rsp->rda);
struct rcu_node *rnp = rcu_get_root(rsp);
if (!rsp->gp_kthread ||
!cpu_needs_another_gp(rsp, rdp)) {
if (!rsp->gp_kthread || !cpu_needs_another_gp(rsp, rdp)) {
/*
* Either we have not yet spawned the grace-period
* task, this CPU does not need another grace period,
* or a grace period is already in progress.
* Either way, don't start a new grace period.
*/
raw_spin_unlock_irqrestore(&rnp->lock, flags);
return;
}
/*
* Because there is no grace period in progress right now,
* any callbacks we have up to this point will be satisfied
* by the next grace period. So this is a good place to
* assign a grace period number to recently posted callbacks.
*/
rcu_accelerate_cbs(rsp, rnp, rdp);
rsp->gp_flags = RCU_GP_FLAG_INIT;
raw_spin_unlock(&rnp->lock); /* Interrupts remain disabled. */
/* Ensure that CPU is aware of completion of last grace period. */
rcu_process_gp_end(rsp, rdp);
local_irq_restore(flags);
/* Wake up rcu_gp_kthread() to start the grace period. */
wake_up(&rsp->gp_wq);
}
/*
* Similar to rcu_start_gp_advanced(), but also advance the calling CPU's
* callbacks. Note that rcu_start_gp_advanced() cannot do this because it
* is invoked indirectly from rcu_advance_cbs(), which would result in
* endless recursion -- or would do so if it wasn't for the self-deadlock
* that is encountered beforehand.
*/
static void
rcu_start_gp(struct rcu_state *rsp)
{
struct rcu_data *rdp = this_cpu_ptr(rsp->rda);
struct rcu_node *rnp = rcu_get_root(rsp);
/*
* If there is no grace period in progress right now, any
* callbacks we have up to this point will be satisfied by the
* next grace period. Also, advancing the callbacks reduces the
* probability of false positives from cpu_needs_another_gp()
* resulting in pointless grace periods. So, advance callbacks
* then start the grace period!
*/
rcu_advance_cbs(rsp, rnp, rdp);
rcu_start_gp_advanced(rsp, rnp, rdp);
}
/*
* Report a full set of quiescent states to the specified rcu_state
* data structure. This involves cleaning up after the prior grace
* period and letting rcu_start_gp() start up the next grace period
* if one is needed. Note that the caller must hold rnp->lock, as
* required by rcu_start_gp(), which will release it.
* if one is needed. Note that the caller must hold rnp->lock, which
* is released before return.
*/
static void rcu_report_qs_rsp(struct rcu_state *rsp, unsigned long flags)
__releases(rcu_get_root(rsp)->lock)
@ -2124,7 +2259,8 @@ __rcu_process_callbacks(struct rcu_state *rsp)
local_irq_save(flags);
if (cpu_needs_another_gp(rsp, rdp)) {
raw_spin_lock(&rcu_get_root(rsp)->lock); /* irqs disabled. */
rcu_start_gp(rsp, flags); /* releases above lock */
rcu_start_gp(rsp);
raw_spin_unlock_irqrestore(&rcu_get_root(rsp)->lock, flags);
} else {
local_irq_restore(flags);
}
@ -2169,7 +2305,8 @@ static void invoke_rcu_callbacks(struct rcu_state *rsp, struct rcu_data *rdp)
static void invoke_rcu_core(void)
{
raise_softirq(RCU_SOFTIRQ);
if (cpu_online(smp_processor_id()))
raise_softirq(RCU_SOFTIRQ);
}
/*
@ -2204,11 +2341,11 @@ static void __call_rcu_core(struct rcu_state *rsp, struct rcu_data *rdp,
/* Start a new grace period if one not already started. */
if (!rcu_gp_in_progress(rsp)) {
unsigned long nestflag;
struct rcu_node *rnp_root = rcu_get_root(rsp);
raw_spin_lock_irqsave(&rnp_root->lock, nestflag);
rcu_start_gp(rsp, nestflag); /* rlses rnp_root->lock */
raw_spin_lock(&rnp_root->lock);
rcu_start_gp(rsp);
raw_spin_unlock(&rnp_root->lock);
} else {
/* Give the grace period a kick. */
rdp->blimit = LONG_MAX;
@ -2628,19 +2765,27 @@ static int rcu_pending(int cpu)
}
/*
* Check to see if any future RCU-related work will need to be done
* by the current CPU, even if none need be done immediately, returning
* 1 if so.
* Return true if the specified CPU has any callback. If all_lazy is
* non-NULL, store an indication of whether all callbacks are lazy.
* (If there are no callbacks, all of them are deemed to be lazy.)
*/
static int rcu_cpu_has_callbacks(int cpu)
static int rcu_cpu_has_callbacks(int cpu, bool *all_lazy)
{
bool al = true;
bool hc = false;
struct rcu_data *rdp;
struct rcu_state *rsp;
/* RCU callbacks either ready or pending? */
for_each_rcu_flavor(rsp)
if (per_cpu_ptr(rsp->rda, cpu)->nxtlist)
return 1;
return 0;
for_each_rcu_flavor(rsp) {
rdp = per_cpu_ptr(rsp->rda, cpu);
if (rdp->qlen != rdp->qlen_lazy)
al = false;
if (rdp->nxtlist)
hc = true;
}
if (all_lazy)
*all_lazy = al;
return hc;
}
/*
@ -2859,7 +3004,6 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptible)
rdp->dynticks->dynticks_nesting = DYNTICK_TASK_EXIT_IDLE;
atomic_set(&rdp->dynticks->dynticks,
(atomic_read(&rdp->dynticks->dynticks) & ~0x1) + 1);
rcu_prepare_for_idle_init(cpu);
raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */
/* Add CPU to rcu_node bitmasks. */
@ -2909,7 +3053,6 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self,
struct rcu_data *rdp = per_cpu_ptr(rcu_state->rda, cpu);
struct rcu_node *rnp = rdp->mynode;
struct rcu_state *rsp;
int ret = NOTIFY_OK;
trace_rcu_utilization("Start CPU hotplug");
switch (action) {
@ -2923,21 +3066,12 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self,
rcu_boost_kthread_setaffinity(rnp, -1);
break;
case CPU_DOWN_PREPARE:
if (nocb_cpu_expendable(cpu))
rcu_boost_kthread_setaffinity(rnp, cpu);
else
ret = NOTIFY_BAD;
rcu_boost_kthread_setaffinity(rnp, cpu);
break;
case CPU_DYING:
case CPU_DYING_FROZEN:
/*
* The whole machine is "stopped" except this CPU, so we can
* touch any data without introducing corruption. We send the
* dying CPU's callbacks to an arbitrarily chosen online CPU.
*/
for_each_rcu_flavor(rsp)
rcu_cleanup_dying_cpu(rsp);
rcu_cleanup_after_idle(cpu);
break;
case CPU_DEAD:
case CPU_DEAD_FROZEN:
@ -2950,7 +3084,7 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self,
break;
}
trace_rcu_utilization("End CPU hotplug");
return ret;
return NOTIFY_OK;
}
/*
@ -3085,6 +3219,7 @@ static void __init rcu_init_one(struct rcu_state *rsp,
}
rnp->level = i;
INIT_LIST_HEAD(&rnp->blkd_tasks);
rcu_init_one_nocb(rnp);
}
}
@ -3170,8 +3305,7 @@ void __init rcu_init(void)
rcu_init_one(&rcu_sched_state, &rcu_sched_data);
rcu_init_one(&rcu_bh_state, &rcu_bh_data);
__rcu_init_preempt();
rcu_init_nocb();
open_softirq(RCU_SOFTIRQ, rcu_process_callbacks);
open_softirq(RCU_SOFTIRQ, rcu_process_callbacks);
/*
* We don't need protection against CPU-hotplug here because

41
kernel/rcutree.h
View File

@ -88,18 +88,13 @@ struct rcu_dynticks {
int dynticks_nmi_nesting; /* Track NMI nesting level. */
atomic_t dynticks; /* Even value for idle, else odd. */
#ifdef CONFIG_RCU_FAST_NO_HZ
int dyntick_drain; /* Prepare-for-idle state variable. */
unsigned long dyntick_holdoff;
/* No retries for the jiffy of failure. */
struct timer_list idle_gp_timer;
/* Wake up CPU sleeping with callbacks. */
unsigned long idle_gp_timer_expires;
/* When to wake up CPU (for repost). */
bool idle_first_pass; /* First pass of attempt to go idle? */
bool all_lazy; /* Are all CPU's CBs lazy? */
unsigned long nonlazy_posted;
/* # times non-lazy CBs posted to CPU. */
unsigned long nonlazy_posted_snap;
/* idle-period nonlazy_posted snapshot. */
unsigned long last_accelerate;
/* Last jiffy CBs were accelerated. */
int tick_nohz_enabled_snap; /* Previously seen value from sysfs. */
#endif /* #ifdef CONFIG_RCU_FAST_NO_HZ */
};
@ -134,9 +129,6 @@ struct rcu_node {
/* elements that need to drain to allow the */
/* current expedited grace period to */
/* complete (only for TREE_PREEMPT_RCU). */
atomic_t wakemask; /* CPUs whose kthread needs to be awakened. */
/* Since this has meaning only for leaf */
/* rcu_node structures, 32 bits suffices. */
unsigned long qsmaskinit;
/* Per-GP initial value for qsmask & expmask. */
unsigned long grpmask; /* Mask to apply to parent qsmask. */
@ -196,6 +188,12 @@ struct rcu_node {
/* Refused to boost: not sure why, though. */
/* This can happen due to race conditions. */
#endif /* #ifdef CONFIG_RCU_BOOST */
#ifdef CONFIG_RCU_NOCB_CPU
wait_queue_head_t nocb_gp_wq[2];
/* Place for rcu_nocb_kthread() to wait GP. */
#endif /* #ifdef CONFIG_RCU_NOCB_CPU */
int need_future_gp[2];
/* Counts of upcoming no-CB GP requests. */
raw_spinlock_t fqslock ____cacheline_internodealigned_in_smp;
} ____cacheline_internodealigned_in_smp;
@ -328,6 +326,11 @@ struct rcu_data {
struct task_struct *nocb_kthread;
#endif /* #ifdef CONFIG_RCU_NOCB_CPU */
/* 8) RCU CPU stall data. */
#ifdef CONFIG_RCU_CPU_STALL_INFO
unsigned int softirq_snap; /* Snapshot of softirq activity. */
#endif /* #ifdef CONFIG_RCU_CPU_STALL_INFO */
int cpu;
struct rcu_state *rsp;
};
@ -375,12 +378,6 @@ struct rcu_state {
struct rcu_data __percpu *rda; /* pointer of percu rcu_data. */
void (*call)(struct rcu_head *head, /* call_rcu() flavor. */
void (*func)(struct rcu_head *head));
#ifdef CONFIG_RCU_NOCB_CPU
void (*call_remote)(struct rcu_head *head,
void (*func)(struct rcu_head *head));
/* call_rcu() flavor, but for */
/* placing on remote CPU. */
#endif /* #ifdef CONFIG_RCU_NOCB_CPU */
/* The following fields are guarded by the root rcu_node's lock. */
@ -443,6 +440,7 @@ struct rcu_state {
unsigned long gp_max; /* Maximum GP duration in */
/* jiffies. */
char *name; /* Name of structure. */
char abbr; /* Abbreviated name. */
struct list_head flavors; /* List of RCU flavors. */
};
@ -520,7 +518,6 @@ static int __cpuinit rcu_spawn_one_boost_kthread(struct rcu_state *rsp,
struct rcu_node *rnp);
#endif /* #ifdef CONFIG_RCU_BOOST */
static void __cpuinit rcu_prepare_kthreads(int cpu);
static void rcu_prepare_for_idle_init(int cpu);
static void rcu_cleanup_after_idle(int cpu);
static void rcu_prepare_for_idle(int cpu);
static void rcu_idle_count_callbacks_posted(void);
@ -529,16 +526,18 @@ static void print_cpu_stall_info(struct rcu_state *rsp, int cpu);
static void print_cpu_stall_info_end(void);
static void zero_cpu_stall_ticks(struct rcu_data *rdp);
static void increment_cpu_stall_ticks(void);
static int rcu_nocb_needs_gp(struct rcu_state *rsp);
static void rcu_nocb_gp_set(struct rcu_node *rnp, int nrq);
static void rcu_nocb_gp_cleanup(struct rcu_state *rsp, struct rcu_node *rnp);
static void rcu_init_one_nocb(struct rcu_node *rnp);
static bool is_nocb_cpu(int cpu);
static bool __call_rcu_nocb(struct rcu_data *rdp, struct rcu_head *rhp,
bool lazy);
static bool rcu_nocb_adopt_orphan_cbs(struct rcu_state *rsp,
struct rcu_data *rdp);
static bool nocb_cpu_expendable(int cpu);
static void rcu_boot_init_nocb_percpu_data(struct rcu_data *rdp);
static void rcu_spawn_nocb_kthreads(struct rcu_state *rsp);
static void init_nocb_callback_list(struct rcu_data *rdp);
static void __init rcu_init_nocb(void);
static bool init_nocb_callback_list(struct rcu_data *rdp);
#endif /* #ifndef RCU_TREE_NONCORE */

623
kernel/rcutree_plugin.h
View File

@ -85,11 +85,21 @@ static void __init rcu_bootup_announce_oddness(void)
if (nr_cpu_ids != NR_CPUS)
printk(KERN_INFO "\tRCU restricting CPUs from NR_CPUS=%d to nr_cpu_ids=%d.\n", NR_CPUS, nr_cpu_ids);
#ifdef CONFIG_RCU_NOCB_CPU
#ifndef CONFIG_RCU_NOCB_CPU_NONE
if (!have_rcu_nocb_mask) {
alloc_bootmem_cpumask_var(&rcu_nocb_mask);
have_rcu_nocb_mask = true;
}
#ifdef CONFIG_RCU_NOCB_CPU_ZERO
pr_info("\tExperimental no-CBs CPU 0\n");
cpumask_set_cpu(0, rcu_nocb_mask);
#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ZERO */
#ifdef CONFIG_RCU_NOCB_CPU_ALL
pr_info("\tExperimental no-CBs for all CPUs\n");
cpumask_setall(rcu_nocb_mask);
#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ALL */
#endif /* #ifndef CONFIG_RCU_NOCB_CPU_NONE */
if (have_rcu_nocb_mask) {
if (cpumask_test_cpu(0, rcu_nocb_mask)) {
cpumask_clear_cpu(0, rcu_nocb_mask);
pr_info("\tCPU 0: illegal no-CBs CPU (cleared).\n");
}
cpulist_scnprintf(nocb_buf, sizeof(nocb_buf), rcu_nocb_mask);
pr_info("\tExperimental no-CBs CPUs: %s.\n", nocb_buf);
if (rcu_nocb_poll)
@ -101,7 +111,7 @@ static void __init rcu_bootup_announce_oddness(void)
#ifdef CONFIG_TREE_PREEMPT_RCU
struct rcu_state rcu_preempt_state =
RCU_STATE_INITIALIZER(rcu_preempt, call_rcu);
RCU_STATE_INITIALIZER(rcu_preempt, 'p', call_rcu);
DEFINE_PER_CPU(struct rcu_data, rcu_preempt_data);
static struct rcu_state *rcu_state = &rcu_preempt_state;
@ -1533,14 +1543,7 @@ static void __cpuinit rcu_prepare_kthreads(int cpu)
int rcu_needs_cpu(int cpu, unsigned long *delta_jiffies)
{
*delta_jiffies = ULONG_MAX;
return rcu_cpu_has_callbacks(cpu);
}
/*
* Because we do not have RCU_FAST_NO_HZ, don't bother initializing for it.
*/
static void rcu_prepare_for_idle_init(int cpu)
{
return rcu_cpu_has_callbacks(cpu, NULL);
}
/*
@ -1577,16 +1580,6 @@ static void rcu_idle_count_callbacks_posted(void)
*
* The following three proprocessor symbols control this state machine:
*
* RCU_IDLE_FLUSHES gives the maximum number of times that we will attempt
* to satisfy RCU. Beyond this point, it is better to incur a periodic
* scheduling-clock interrupt than to loop through the state machine
* at full power.
* RCU_IDLE_OPT_FLUSHES gives the number of RCU_IDLE_FLUSHES that are
* optional if RCU does not need anything immediately from this
* CPU, even if this CPU still has RCU callbacks queued. The first
* times through the state machine are mandatory: we need to give
* the state machine a chance to communicate a quiescent state
* to the RCU core.
* RCU_IDLE_GP_DELAY gives the number of jiffies that a CPU is permitted
* to sleep in dyntick-idle mode with RCU callbacks pending. This
* is sized to be roughly one RCU grace period. Those energy-efficiency
@ -1602,186 +1595,108 @@ static void rcu_idle_count_callbacks_posted(void)
* adjustment, they can be converted into kernel config parameters, though
* making the state machine smarter might be a better option.
*/
#define RCU_IDLE_FLUSHES 5 /* Number of dyntick-idle tries. */
#define RCU_IDLE_OPT_FLUSHES 3 /* Optional dyntick-idle tries. */
#define RCU_IDLE_GP_DELAY 4 /* Roughly one grace period. */
#define RCU_IDLE_LAZY_GP_DELAY (6 * HZ) /* Roughly six seconds. */
static int rcu_idle_gp_delay = RCU_IDLE_GP_DELAY;
module_param(rcu_idle_gp_delay, int, 0644);
static int rcu_idle_lazy_gp_delay = RCU_IDLE_LAZY_GP_DELAY;
module_param(rcu_idle_lazy_gp_delay, int, 0644);
extern int tick_nohz_enabled;
/*
* Does the specified flavor of RCU have non-lazy callbacks pending on
* the specified CPU? Both RCU flavor and CPU are specified by the
* rcu_data structure.
* Try to advance callbacks for all flavors of RCU on the current CPU.
* Afterwards, if there are any callbacks ready for immediate invocation,
* return true.
*/
static bool __rcu_cpu_has_nonlazy_callbacks(struct rcu_data *rdp)
static bool rcu_try_advance_all_cbs(void)
{
return rdp->qlen != rdp->qlen_lazy;
}
bool cbs_ready = false;
struct rcu_data *rdp;
struct rcu_node *rnp;
struct rcu_state *rsp;
#ifdef CONFIG_TREE_PREEMPT_RCU
for_each_rcu_flavor(rsp) {
rdp = this_cpu_ptr(rsp->rda);
rnp = rdp->mynode;
/*
* Are there non-lazy RCU-preempt callbacks? (There cannot be if there
* is no RCU-preempt in the kernel.)
*/
static bool rcu_preempt_cpu_has_nonlazy_callbacks(int cpu)
{
struct rcu_data *rdp = &per_cpu(rcu_preempt_data, cpu);
/*
* Don't bother checking unless a grace period has
* completed since we last checked and there are
* callbacks not yet ready to invoke.
*/
if (rdp->completed != rnp->completed &&
rdp->nxttail[RCU_DONE_TAIL] != rdp->nxttail[RCU_NEXT_TAIL])
rcu_process_gp_end(rsp, rdp);
return __rcu_cpu_has_nonlazy_callbacks(rdp);
}
#else /* #ifdef CONFIG_TREE_PREEMPT_RCU */
static bool rcu_preempt_cpu_has_nonlazy_callbacks(int cpu)
{
return 0;
}
#endif /* else #ifdef CONFIG_TREE_PREEMPT_RCU */
/*
* Does any flavor of RCU have non-lazy callbacks on the specified CPU?
*/
static bool rcu_cpu_has_nonlazy_callbacks(int cpu)
{
return __rcu_cpu_has_nonlazy_callbacks(&per_cpu(rcu_sched_data, cpu)) ||
__rcu_cpu_has_nonlazy_callbacks(&per_cpu(rcu_bh_data, cpu)) ||
rcu_preempt_cpu_has_nonlazy_callbacks(cpu);
if (cpu_has_callbacks_ready_to_invoke(rdp))
cbs_ready = true;
}
return cbs_ready;
}
/*
* Allow the CPU to enter dyntick-idle mode if either: (1) There are no
* callbacks on this CPU, (2) this CPU has not yet attempted to enter
* dyntick-idle mode, or (3) this CPU is in the process of attempting to
* enter dyntick-idle mode. Otherwise, if we have recently tried and failed
* to enter dyntick-idle mode, we refuse to try to enter it. After all,
* it is better to incur scheduling-clock interrupts than to spin
* continuously for the same time duration!
* Allow the CPU to enter dyntick-idle mode unless it has callbacks ready
* to invoke. If the CPU has callbacks, try to advance them. Tell the
* caller to set the timeout based on whether or not there are non-lazy
* callbacks.
*
* The delta_jiffies argument is used to store the time when RCU is
* going to need the CPU again if it still has callbacks. The reason
* for this is that rcu_prepare_for_idle() might need to post a timer,
* but if so, it will do so after tick_nohz_stop_sched_tick() has set
* the wakeup time for this CPU. This means that RCU's timer can be
* delayed until the wakeup time, which defeats the purpose of posting
* a timer.
* The caller must have disabled interrupts.
*/
int rcu_needs_cpu(int cpu, unsigned long *delta_jiffies)
int rcu_needs_cpu(int cpu, unsigned long *dj)
{
struct rcu_dynticks *rdtp = &per_cpu(rcu_dynticks, cpu);
/* Flag a new idle sojourn to the idle-entry state machine. */
rdtp->idle_first_pass = 1;
/* Snapshot to detect later posting of non-lazy callback. */
rdtp->nonlazy_posted_snap = rdtp->nonlazy_posted;
/* If no callbacks, RCU doesn't need the CPU. */
if (!rcu_cpu_has_callbacks(cpu)) {
*delta_jiffies = ULONG_MAX;
if (!rcu_cpu_has_callbacks(cpu, &rdtp->all_lazy)) {
*dj = ULONG_MAX;
return 0;
}
if (rdtp->dyntick_holdoff == jiffies) {
/* RCU recently tried and failed, so don't try again. */
*delta_jiffies = 1;
/* Attempt to advance callbacks. */
if (rcu_try_advance_all_cbs()) {
/* Some ready to invoke, so initiate later invocation. */
invoke_rcu_core();
return 1;
}
/* Set up for the possibility that RCU will post a timer. */
if (rcu_cpu_has_nonlazy_callbacks(cpu)) {
*delta_jiffies = round_up(RCU_IDLE_GP_DELAY + jiffies,
RCU_IDLE_GP_DELAY) - jiffies;
rdtp->last_accelerate = jiffies;
/* Request timer delay depending on laziness, and round. */
if (rdtp->all_lazy) {
*dj = round_up(rcu_idle_gp_delay + jiffies,
rcu_idle_gp_delay) - jiffies;
} else {
*delta_jiffies = jiffies + RCU_IDLE_LAZY_GP_DELAY;
*delta_jiffies = round_jiffies(*delta_jiffies) - jiffies;
*dj = round_jiffies(rcu_idle_lazy_gp_delay + jiffies) - jiffies;
}
return 0;
}
/*
* Handler for smp_call_function_single(). The only point of this
* handler is to wake the CPU up, so the handler does only tracing.
*/
void rcu_idle_demigrate(void *unused)
{
trace_rcu_prep_idle("Demigrate");
}
/*
* Timer handler used to force CPU to start pushing its remaining RCU
* callbacks in the case where it entered dyntick-idle mode with callbacks
* pending. The hander doesn't really need to do anything because the
* real work is done upon re-entry to idle, or by the next scheduling-clock
* interrupt should idle not be re-entered.
*
* One special case: the timer gets migrated without awakening the CPU
* on which the timer was scheduled on. In this case, we must wake up
* that CPU. We do so with smp_call_function_single().
*/
static void rcu_idle_gp_timer_func(unsigned long cpu_in)
{
int cpu = (int)cpu_in;
trace_rcu_prep_idle("Timer");
if (cpu != smp_processor_id())
smp_call_function_single(cpu, rcu_idle_demigrate, NULL, 0);
else
WARN_ON_ONCE(1); /* Getting here can hang the system... */
}
/*
* Initialize the timer used to pull CPUs out of dyntick-idle mode.
*/
static void rcu_prepare_for_idle_init(int cpu)
{
struct rcu_dynticks *rdtp = &per_cpu(rcu_dynticks, cpu);
rdtp->dyntick_holdoff = jiffies - 1;
setup_timer(&rdtp->idle_gp_timer, rcu_idle_gp_timer_func, cpu);
rdtp->idle_gp_timer_expires = jiffies - 1;
rdtp->idle_first_pass = 1;
}
/*
* Clean up for exit from idle. Because we are exiting from idle, there
* is no longer any point to ->idle_gp_timer, so cancel it. This will
* do nothing if this timer is not active, so just cancel it unconditionally.
*/
static void rcu_cleanup_after_idle(int cpu)
{
struct rcu_dynticks *rdtp = &per_cpu(rcu_dynticks, cpu);
del_timer(&rdtp->idle_gp_timer);
trace_rcu_prep_idle("Cleanup after idle");
rdtp->tick_nohz_enabled_snap = ACCESS_ONCE(tick_nohz_enabled);
}
/*
* Check to see if any RCU-related work can be done by the current CPU,
* and if so, schedule a softirq to get it done. This function is part
* of the RCU implementation; it is -not- an exported member of the RCU API.
*
* The idea is for the current CPU to clear out all work required by the
* RCU core for the current grace period, so that this CPU can be permitted
* to enter dyntick-idle mode. In some cases, it will need to be awakened
* at the end of the grace period by whatever CPU ends the grace period.
* This allows CPUs to go dyntick-idle more quickly, and to reduce the
* number of wakeups by a modest integer factor.
*
* Because it is not legal to invoke rcu_process_callbacks() with irqs
* disabled, we do one pass of force_quiescent_state(), then do a
* invoke_rcu_core() to cause rcu_process_callbacks() to be invoked
* later. The ->dyntick_drain field controls the sequencing.
* Prepare a CPU for idle from an RCU perspective. The first major task
* is to sense whether nohz mode has been enabled or disabled via sysfs.
* The second major task is to check to see if a non-lazy callback has
* arrived at a CPU that previously had only lazy callbacks. The third
* major task is to accelerate (that is, assign grace-period numbers to)
* any recently arrived callbacks.
*
* The caller must have disabled interrupts.
*/
static void rcu_prepare_for_idle(int cpu)
{
struct timer_list *tp;
struct rcu_data *rdp;
struct rcu_dynticks *rdtp = &per_cpu(rcu_dynticks, cpu);
struct rcu_node *rnp;
struct rcu_state *rsp;
int tne;
/* Handle nohz enablement switches conservatively. */
tne = ACCESS_ONCE(tick_nohz_enabled);
if (tne != rdtp->tick_nohz_enabled_snap) {
if (rcu_cpu_has_callbacks(cpu))
if (rcu_cpu_has_callbacks(cpu, NULL))
invoke_rcu_core(); /* force nohz to see update. */
rdtp->tick_nohz_enabled_snap = tne;
return;
@ -1789,125 +1704,56 @@ static void rcu_prepare_for_idle(int cpu)
if (!tne)
return;
/* Adaptive-tick mode, where usermode execution is idle to RCU. */
if (!is_idle_task(current)) {
rdtp->dyntick_holdoff = jiffies - 1;
if (rcu_cpu_has_nonlazy_callbacks(cpu)) {
trace_rcu_prep_idle("User dyntick with callbacks");
rdtp->idle_gp_timer_expires =
round_up(jiffies + RCU_IDLE_GP_DELAY,
RCU_IDLE_GP_DELAY);
} else if (rcu_cpu_has_callbacks(cpu)) {
rdtp->idle_gp_timer_expires =
round_jiffies(jiffies + RCU_IDLE_LAZY_GP_DELAY);
trace_rcu_prep_idle("User dyntick with lazy callbacks");
} else {
return;
}
tp = &rdtp->idle_gp_timer;
mod_timer_pinned(tp, rdtp->idle_gp_timer_expires);
/* If this is a no-CBs CPU, no callbacks, just return. */
if (is_nocb_cpu(cpu))
return;
}
/*
* If this is an idle re-entry, for example, due to use of
* RCU_NONIDLE() or the new idle-loop tracing API within the idle
* loop, then don't take any state-machine actions, unless the
* momentary exit from idle queued additional non-lazy callbacks.
* Instead, repost the ->idle_gp_timer if this CPU has callbacks
* pending.
* If a non-lazy callback arrived at a CPU having only lazy
* callbacks, invoke RCU core for the side-effect of recalculating
* idle duration on re-entry to idle.
*/
if (!rdtp->idle_first_pass &&
(rdtp->nonlazy_posted == rdtp->nonlazy_posted_snap)) {
if (rcu_cpu_has_callbacks(cpu)) {
tp = &rdtp->idle_gp_timer;
mod_timer_pinned(tp, rdtp->idle_gp_timer_expires);
}
return;
}
rdtp->idle_first_pass = 0;
rdtp->nonlazy_posted_snap = rdtp->nonlazy_posted - 1;
/*
* If there are no callbacks on this CPU, enter dyntick-idle mode.
* Also reset state to avoid prejudicing later attempts.
*/
if (!rcu_cpu_has_callbacks(cpu)) {
rdtp->dyntick_holdoff = jiffies - 1;
rdtp->dyntick_drain = 0;
trace_rcu_prep_idle("No callbacks");
return;
}
/*
* If in holdoff mode, just return. We will presumably have
* refrained from disabling the scheduling-clock tick.
*/
if (rdtp->dyntick_holdoff == jiffies) {
trace_rcu_prep_idle("In holdoff");
return;
}
/* Check and update the ->dyntick_drain sequencing. */
if (rdtp->dyntick_drain <= 0) {
/* First time through, initialize the counter. */
rdtp->dyntick_drain = RCU_IDLE_FLUSHES;
} else if (rdtp->dyntick_drain <= RCU_IDLE_OPT_FLUSHES &&
!rcu_pending(cpu) &&
!local_softirq_pending()) {
/* Can we go dyntick-idle despite still having callbacks? */
rdtp->dyntick_drain = 0;
rdtp->dyntick_holdoff = jiffies;
if (rcu_cpu_has_nonlazy_callbacks(cpu)) {
trace_rcu_prep_idle("Dyntick with callbacks");
rdtp->idle_gp_timer_expires =
round_up(jiffies + RCU_IDLE_GP_DELAY,
RCU_IDLE_GP_DELAY);
} else {
rdtp->idle_gp_timer_expires =
round_jiffies(jiffies + RCU_IDLE_LAZY_GP_DELAY);
trace_rcu_prep_idle("Dyntick with lazy callbacks");
}
tp = &rdtp->idle_gp_timer;
mod_timer_pinned(tp, rdtp->idle_gp_timer_expires);
rdtp->nonlazy_posted_snap = rdtp->nonlazy_posted;
return; /* Nothing more to do immediately. */
} else if (--(rdtp->dyntick_drain) <= 0) {
/* We have hit the limit, so time to give up. */
rdtp->dyntick_holdoff = jiffies;
trace_rcu_prep_idle("Begin holdoff");
invoke_rcu_core(); /* Force the CPU out of dyntick-idle. */
return;
}
/*
* Do one step of pushing the remaining RCU callbacks through
* the RCU core state machine.
*/
#ifdef CONFIG_TREE_PREEMPT_RCU
if (per_cpu(rcu_preempt_data, cpu).nxtlist) {
rcu_preempt_qs(cpu);
force_quiescent_state(&rcu_preempt_state);
}
#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */
if (per_cpu(rcu_sched_data, cpu).nxtlist) {
rcu_sched_qs(cpu);
force_quiescent_state(&rcu_sched_state);
}
if (per_cpu(rcu_bh_data, cpu).nxtlist) {
rcu_bh_qs(cpu);
force_quiescent_state(&rcu_bh_state);
}
/*
* If RCU callbacks are still pending, RCU still needs this CPU.
* So try forcing the callbacks through the grace period.
*/
if (rcu_cpu_has_callbacks(cpu)) {
trace_rcu_prep_idle("More callbacks");
if (rdtp->all_lazy &&
rdtp->nonlazy_posted != rdtp->nonlazy_posted_snap) {
invoke_rcu_core();
} else {
trace_rcu_prep_idle("Callbacks drained");
return;
}
/*
* If we have not yet accelerated this jiffy, accelerate all
* callbacks on this CPU.
*/
if (rdtp->last_accelerate == jiffies)
return;
rdtp->last_accelerate = jiffies;
for_each_rcu_flavor(rsp) {
rdp = per_cpu_ptr(rsp->rda, cpu);
if (!*rdp->nxttail[RCU_DONE_TAIL])
continue;
rnp = rdp->mynode;
raw_spin_lock(&rnp->lock); /* irqs already disabled. */
rcu_accelerate_cbs(rsp, rnp, rdp);
raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */
}
}
/*
* Clean up for exit from idle. Attempt to advance callbacks based on
* any grace periods that elapsed while the CPU was idle, and if any
* callbacks are now ready to invoke, initiate invocation.
*/
static void rcu_cleanup_after_idle(int cpu)
{
struct rcu_data *rdp;
struct rcu_state *rsp;
if (is_nocb_cpu(cpu))
return;
rcu_try_advance_all_cbs();
for_each_rcu_flavor(rsp) {
rdp = per_cpu_ptr(rsp->rda, cpu);
if (cpu_has_callbacks_ready_to_invoke(rdp))
invoke_rcu_core();
}
}
@ -2015,16 +1861,13 @@ early_initcall(rcu_register_oom_notifier);
static void print_cpu_stall_fast_no_hz(char *cp, int cpu)
{
struct rcu_dynticks *rdtp = &per_cpu(rcu_dynticks, cpu);
struct timer_list *tltp = &rdtp->idle_gp_timer;
char c;
unsigned long nlpd = rdtp->nonlazy_posted - rdtp->nonlazy_posted_snap;
c = rdtp->dyntick_holdoff == jiffies ? 'H' : '.';
if (timer_pending(tltp))
sprintf(cp, "drain=%d %c timer=%lu",
rdtp->dyntick_drain, c, tltp->expires - jiffies);
else
sprintf(cp, "drain=%d %c timer not pending",
rdtp->dyntick_drain, c);
sprintf(cp, "last_accelerate: %04lx/%04lx, nonlazy_posted: %ld, %c%c",
rdtp->last_accelerate & 0xffff, jiffies & 0xffff,
ulong2long(nlpd),
rdtp->all_lazy ? 'L' : '.',
rdtp->tick_nohz_enabled_snap ? '.' : 'D');
}
#else /* #ifdef CONFIG_RCU_FAST_NO_HZ */
@ -2070,10 +1913,11 @@ static void print_cpu_stall_info(struct rcu_state *rsp, int cpu)
ticks_value = rsp->gpnum - rdp->gpnum;
}
print_cpu_stall_fast_no_hz(fast_no_hz, cpu);
printk(KERN_ERR "\t%d: (%lu %s) idle=%03x/%llx/%d %s\n",
printk(KERN_ERR "\t%d: (%lu %s) idle=%03x/%llx/%d softirq=%u/%u %s\n",
cpu, ticks_value, ticks_title,
atomic_read(&rdtp->dynticks) & 0xfff,
rdtp->dynticks_nesting, rdtp->dynticks_nmi_nesting,
rdp->softirq_snap, kstat_softirqs_cpu(RCU_SOFTIRQ, cpu),
fast_no_hz);
}
@ -2087,6 +1931,7 @@ static void print_cpu_stall_info_end(void)
static void zero_cpu_stall_ticks(struct rcu_data *rdp)
{
rdp->ticks_this_gp = 0;
rdp->softirq_snap = kstat_softirqs_cpu(RCU_SOFTIRQ, smp_processor_id());
}
/* Increment ->ticks_this_gp for all flavors of RCU. */
@ -2165,6 +2010,47 @@ static int __init parse_rcu_nocb_poll(char *arg)
}
early_param("rcu_nocb_poll", parse_rcu_nocb_poll);
/*
* Do any no-CBs CPUs need another grace period?
*
* Interrupts must be disabled. If the caller does not hold the root
* rnp_node structure's ->lock, the results are advisory only.
*/
static int rcu_nocb_needs_gp(struct rcu_state *rsp)
{
struct rcu_node *rnp = rcu_get_root(rsp);
return rnp->need_future_gp[(ACCESS_ONCE(rnp->completed) + 1) & 0x1];
}
/*
* Wake up any no-CBs CPUs' kthreads that were waiting on the just-ended
* grace period.
*/
static void rcu_nocb_gp_cleanup(struct rcu_state *rsp, struct rcu_node *rnp)
{
wake_up_all(&rnp->nocb_gp_wq[rnp->completed & 0x1]);
}
/*
* Set the root rcu_node structure's ->need_future_gp field
* based on the sum of those of all rcu_node structures. This does
* double-count the root rcu_node structure's requests, but this
* is necessary to handle the possibility of a rcu_nocb_kthread()
* having awakened during the time that the rcu_node structures
* were being updated for the end of the previous grace period.
*/
static void rcu_nocb_gp_set(struct rcu_node *rnp, int nrq)
{
rnp->need_future_gp[(rnp->completed + 1) & 0x1] += nrq;
}
static void rcu_init_one_nocb(struct rcu_node *rnp)
{
init_waitqueue_head(&rnp->nocb_gp_wq[0]);
init_waitqueue_head(&rnp->nocb_gp_wq[1]);
}
/* Is the specified CPU a no-CPUs CPU? */
static bool is_nocb_cpu(int cpu)
{
@ -2227,6 +2113,13 @@ static bool __call_rcu_nocb(struct rcu_data *rdp, struct rcu_head *rhp,
if (!is_nocb_cpu(rdp->cpu))
return 0;
__call_rcu_nocb_enqueue(rdp, rhp, &rhp->next, 1, lazy);
if (__is_kfree_rcu_offset((unsigned long)rhp->func))
trace_rcu_kfree_callback(rdp->rsp->name, rhp,
(unsigned long)rhp->func,
rdp->qlen_lazy, rdp->qlen);
else
trace_rcu_callback(rdp->rsp->name, rhp,
rdp->qlen_lazy, rdp->qlen);
return 1;
}
@ -2265,95 +2158,36 @@ static bool __maybe_unused rcu_nocb_adopt_orphan_cbs(struct rcu_state *rsp,
}
/*
* There must be at least one non-no-CBs CPU in operation at any given
* time, because no-CBs CPUs are not capable of initiating grace periods
* independently. This function therefore complains if the specified
* CPU is the last non-no-CBs CPU, allowing the CPU-hotplug system to
* avoid offlining the last such CPU. (Recursion is a wonderful thing,
* but you have to have a base case!)
* If necessary, kick off a new grace period, and either way wait
* for a subsequent grace period to complete.
*/
static bool nocb_cpu_expendable(int cpu)
static void rcu_nocb_wait_gp(struct rcu_data *rdp)
{
cpumask_var_t non_nocb_cpus;
int ret;
unsigned long c;
bool d;
unsigned long flags;
struct rcu_node *rnp = rdp->mynode;
raw_spin_lock_irqsave(&rnp->lock, flags);
c = rcu_start_future_gp(rnp, rdp);
raw_spin_unlock_irqrestore(&rnp->lock, flags);
/*
* If there are no no-CB CPUs or if this CPU is not a no-CB CPU,
* then offlining this CPU is harmless. Let it happen.
* Wait for the grace period. Do so interruptibly to avoid messing
* up the load average.
*/
if (!have_rcu_nocb_mask || is_nocb_cpu(cpu))
return 1;
/* If no memory, play it safe and keep the CPU around. */
if (!alloc_cpumask_var(&non_nocb_cpus, GFP_NOIO))
return 0;
cpumask_andnot(non_nocb_cpus, cpu_online_mask, rcu_nocb_mask);
cpumask_clear_cpu(cpu, non_nocb_cpus);
ret = !cpumask_empty(non_nocb_cpus);
free_cpumask_var(non_nocb_cpus);
return ret;
}
/*
* Helper structure for remote registry of RCU callbacks.
* This is needed for when a no-CBs CPU needs to start a grace period.
* If it just invokes call_rcu(), the resulting callback will be queued,
* which can result in deadlock.
*/
struct rcu_head_remote {
struct rcu_head *rhp;
call_rcu_func_t *crf;
void (*func)(struct rcu_head *rhp);
};
/*
* Register a callback as specified by the rcu_head_remote struct.
* This function is intended to be invoked via smp_call_function_single().
*/
static void call_rcu_local(void *arg)
{
struct rcu_head_remote *rhrp =
container_of(arg, struct rcu_head_remote, rhp);
rhrp->crf(rhrp->rhp, rhrp->func);
}
/*
* Set up an rcu_head_remote structure and the invoke call_rcu_local()
* on CPU 0 (which is guaranteed to be a non-no-CBs CPU) via
* smp_call_function_single().
*/
static void invoke_crf_remote(struct rcu_head *rhp,
void (*func)(struct rcu_head *rhp),
call_rcu_func_t crf)
{
struct rcu_head_remote rhr;
rhr.rhp = rhp;
rhr.crf = crf;
rhr.func = func;
smp_call_function_single(0, call_rcu_local, &rhr, 1);
}
/*
* Helper functions to be passed to wait_rcu_gp(), each of which
* invokes invoke_crf_remote() to register a callback appropriately.
*/
static void __maybe_unused
call_rcu_preempt_remote(struct rcu_head *rhp,
void (*func)(struct rcu_head *rhp))
{
invoke_crf_remote(rhp, func, call_rcu);
}
static void call_rcu_bh_remote(struct rcu_head *rhp,
void (*func)(struct rcu_head *rhp))
{
invoke_crf_remote(rhp, func, call_rcu_bh);
}
static void call_rcu_sched_remote(struct rcu_head *rhp,
void (*func)(struct rcu_head *rhp))
{
invoke_crf_remote(rhp, func, call_rcu_sched);
trace_rcu_future_gp(rnp, rdp, c, "StartWait");
for (;;) {
wait_event_interruptible(
rnp->nocb_gp_wq[c & 0x1],
(d = ULONG_CMP_GE(ACCESS_ONCE(rnp->completed), c)));
if (likely(d))
break;
flush_signals(current);
trace_rcu_future_gp(rnp, rdp, c, "ResumeWait");
}
trace_rcu_future_gp(rnp, rdp, c, "EndWait");
smp_mb(); /* Ensure that CB invocation happens after GP end. */
}
/*
@ -2390,7 +2224,7 @@ static int rcu_nocb_kthread(void *arg)
cl = atomic_long_xchg(&rdp->nocb_q_count_lazy, 0);
ACCESS_ONCE(rdp->nocb_p_count) += c;
ACCESS_ONCE(rdp->nocb_p_count_lazy) += cl;
wait_rcu_gp(rdp->rsp->call_remote);
rcu_nocb_wait_gp(rdp);
/* Each pass through the following loop invokes a callback. */
trace_rcu_batch_start(rdp->rsp->name, cl, c, -1);
@ -2436,33 +2270,42 @@ static void __init rcu_spawn_nocb_kthreads(struct rcu_state *rsp)
return;
for_each_cpu(cpu, rcu_nocb_mask) {
rdp = per_cpu_ptr(rsp->rda, cpu);
t = kthread_run(rcu_nocb_kthread, rdp, "rcuo%d", cpu);
t = kthread_run(rcu_nocb_kthread, rdp,
"rcuo%c/%d", rsp->abbr, cpu);
BUG_ON(IS_ERR(t));
ACCESS_ONCE(rdp->nocb_kthread) = t;
}
}
/* Prevent __call_rcu() from enqueuing callbacks on no-CBs CPUs */
static void init_nocb_callback_list(struct rcu_data *rdp)
static bool init_nocb_callback_list(struct rcu_data *rdp)
{
if (rcu_nocb_mask == NULL ||
!cpumask_test_cpu(rdp->cpu, rcu_nocb_mask))
return;
return false;
rdp->nxttail[RCU_NEXT_TAIL] = NULL;
}
/* Initialize the ->call_remote fields in the rcu_state structures. */
static void __init rcu_init_nocb(void)
{
#ifdef CONFIG_PREEMPT_RCU
rcu_preempt_state.call_remote = call_rcu_preempt_remote;
#endif /* #ifdef CONFIG_PREEMPT_RCU */
rcu_bh_state.call_remote = call_rcu_bh_remote;
rcu_sched_state.call_remote = call_rcu_sched_remote;
return true;
}
#else /* #ifdef CONFIG_RCU_NOCB_CPU */
static int rcu_nocb_needs_gp(struct rcu_state *rsp)
{
return 0;
}
static void rcu_nocb_gp_cleanup(struct rcu_state *rsp, struct rcu_node *rnp)
{
}
static void rcu_nocb_gp_set(struct rcu_node *rnp, int nrq)
{
}
static void rcu_init_one_nocb(struct rcu_node *rnp)
{
}
static bool is_nocb_cpu(int cpu)
{
return false;
@ -2480,11 +2323,6 @@ static bool __maybe_unused rcu_nocb_adopt_orphan_cbs(struct rcu_state *rsp,
return 0;
}
static bool nocb_cpu_expendable(int cpu)
{
return 1;
}
static void __init rcu_boot_init_nocb_percpu_data(struct rcu_data *rdp)
{
}
@ -2493,12 +2331,9 @@ static void __init rcu_spawn_nocb_kthreads(struct rcu_state *rsp)
{
}
static void init_nocb_callback_list(struct rcu_data *rdp)
{
}
static void __init rcu_init_nocb(void)
static bool init_nocb_callback_list(struct rcu_data *rdp)
{
return false;
}
#endif /* #else #ifdef CONFIG_RCU_NOCB_CPU */

2
kernel/rcutree_trace.c
View File

@ -46,8 +46,6 @@
#define RCU_TREE_NONCORE
#include "rcutree.h"
#define ulong2long(a) (*(long *)(&(a)))
static int r_open(struct inode *inode, struct file *file,
const struct seq_operations *op)
{