This patchset introduces wakeup hints for some of the most popular (from
epoll POV) devices, so that epoll code can avoid spurious wakeups on its
waiters.
The problem with epoll is that the callback-based wakeups do not, ATM,
carry any information about the events the wakeup is related to. So the
only choice epoll has (not being able to call f_op->poll() from inside the
callback), is to add the file* to a ready-list and resolve the real events
later on, at epoll_wait() (or its own f_op->poll()) time. This can cause
spurious wakeups, since the wake_up() itself might be for an event the
caller is not interested into.
The rate of these spurious wakeup can be pretty high in case of many
network sockets being monitored.
By allowing devices to report the events the wakeups refer to (at least
the two major classes - POLLIN/POLLOUT), we are able to spare useless
wakeups by proper handling inside the epoll's poll callback.
Epoll will have in any case to call f_op->poll() on the file* later on,
since the change to be done in order to have the full event set sent via
wakeup, is too invasive for the way our f_op->poll() system works (the
full event set is calculated inside the poll function - there are too many
of them to even start thinking the change - also poll/select would need
change too).
Epoll is changed in a way that both devices which send event hints, and
the ones that don't, are correctly handled. The former will gain some
efficiency though.
As a general rule for devices, would be to add an event mask by using
key-aware wakeup macros, when making up poll wait queues. I tested it
(together with the epoll's poll fix patch Andrew has in -mm) and wakeups
for the supported devices are correctly filtered.
Test program available here:
http://www.xmailserver.org/epoll_test.c
This patch:
Nothing revolutionary here. Just using the available "key" that our
wakeup core already support. The __wake_up_locked_key() was no brainer,
since both __wake_up_locked() and __wake_up_locked_key() are thin wrappers
around __wake_up_common().
The __wake_up_sync() function had a body, so the choice was between
borrowing the body for __wake_up_sync_key() and calling it from
__wake_up_sync(), or make an inline and calling it from both. I chose the
former since in most archs it all resolves to "mov $0, REG; jmp ADDR".
Signed-off-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Cc: William Lee Irwin III <wli@movementarian.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Impact: extend debug info /proc/sched_debug
If the user changes the value of the sched_mc/smt_power_savings sysfs
tunable, it'll trigger a rebuilding of the whole sched_domain tree,
with the SD_POWERSAVINGS_BALANCE flag set at certain levels.
As a result, there would be a change in the __cpu_power of sched_groups
in the sched_domain hierarchy.
Print the __cpu_power values for each sched_group in sched_domain_debug
to help verify this change and correlate it with the change in the
load-balancing behavior.
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20090330045520.2869.24777.stgit@sofia.in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Add per-cgroup cpuacct controller statistics like the system and user
time consumed by the group of tasks.
Changelog:
v7
- Changed the name of the statistic from utime to user and from stime to
system so that in future we could easily add other statistics like irq,
softirq, steal times etc easily.
v6
- Fixed a bug in the error path of cpuacct_create() (pointed by Li Zefan).
v5
- In cpuacct_stats_show(), use cputime64_to_clock_t() since we are
operating on a 64bit variable here.
v4
- Remove comments in cpuacct_update_stats() which explained why rcu_read_lock()
was needed (as per Peter Zijlstra's review comments).
- Don't say that percpu_counter_read() is broken in Documentation/cpuacct.txt
as per KAMEZAWA Hiroyuki's review comments.
v3
- Fix a small race in the cpuacct hierarchy walk.
v2
- stime and utime now exported in clock_t units instead of msecs.
- Addressed the code review comments from Balbir and Li Zefan.
- Moved to -tip tree.
v1
- Moved the stime/utime accounting to cpuacct controller.
Earlier versions
- http://lkml.org/lkml/2009/2/25/129
Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
Signed-off-by: Balaji Rao <balajirrao@gmail.com>
Cc: Dhaval Giani <dhaval@linux.vnet.ibm.com>
Cc: Paul Menage <menage@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Tested-by: Balbir Singh <balbir@linux.vnet.ibm.com>
LKML-Reference: <20090331043222.GA4093@in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: Regression fix (against clock_gettime() backwarding bug)
This patch re-introduces a couple of functions, task_sched_runtime
and thread_group_sched_runtime, which was once removed at the
time of 2.6.28-rc1.
These functions protect the sampling of thread/process clock with
rq lock. This rq lock is required not to update rq->clock during
the sampling.
i.e.
The clock_gettime() may return
((accounted runtime before update) + (delta after update))
that is less than what it should be.
v2 -> v3:
- Rename static helper function __task_delta_exec()
to do_task_delta_exec() since -tip tree already has
a __task_delta_exec() of different version.
v1 -> v2:
- Revises comments of function and patch description.
- Add note about accuracy of thread group's runtime.
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: stable@kernel.org [2.6.28.x][2.6.29.x]
LKML-Reference: <49D1CC93.4080401@jp.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix cgroups race under rcu-preempt
cpuacct_charge() obtains task's ca and does a hierarchy walk upwards.
This can race with the task's movement between cgroups. This race
can cause an access to freed ca pointer in cpuacct_charge() or access
to invalid cgroups pointer of the task. This will not happen with rcu or
tree rcu as cpuacct_charge() is called with preemption disabled. However if
rcupreempt is used, the race is seen. Thanks to Li Zefan for explaining this.
Fix this race by explicitly protecting ca and the hierarchy walk with
rcu_read_lock().
Changes for v2:
- Update patch descrition (as per Li Zefan's review comments).
- Remove comments in cpuacct_charge() which explained why rcu_read_lock()
was needed (as per Peter Zijlstra's review comments).
Signed-off-by: Bharata B Rao <bharata@linux.vnet.ibm.com>
Cc: Dhaval Giani <dhaval@linux.vnet.ibm.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Paul Menage <menage@google.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Tested-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
It appears I inadvertly introduced rq->lock recursion to the
hrtimer_start() path when I delegated running already expired
timers to softirq context.
This patch fixes it by introducing a __hrtimer_start_range_ns()
method that will not use raise_softirq_irqoff() but
__raise_softirq_irqoff() which avoids the wakeup.
It then also changes schedule() to check for pending softirqs and
do the wakeup then, I'm not quite sure I like this last bit, nor
am I convinced its really needed.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: paulus@samba.org
LKML-Reference: <20090313112301.096138802@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This reverts commit 7a46c594bf.
This was applied to the x86 tree mistakenly, it belongs into the
scheduler tree.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Fix kernel-doc errors in sched.c: the structs don't have
kernel-doc notation and the short function description needs to
be one line only.
Error(kernel/sched.c:3197): cannot understand prototype: 'struct sd_lb_stats '
Error(kernel/sched.c:3228): cannot understand prototype: 'struct sg_lb_stats '
Error(kernel/sched.c:3375): duplicate section name 'Description'
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Impact: cleanup
Add /** style comments around find_busiest_group(). Also add a few
explanatory comments.
This concludes the find_busiest_group() cleanup. The function is
now down to 72 lines from the original 313 lines.
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Balbir Singh" <balbir@in.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: "Dhaval Giani" <dhaval@linux.vnet.ibm.com>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
Cc: "Vaidyanathan Srinivasan" <svaidy@linux.vnet.ibm.com>
LKML-Reference: <20090325091427.13992.18933.stgit@sofia.in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
Create seperate helper functions to initialize the
power-savings-balance related variables, to update them and
to check if we have a scope for performing power-savings balance.
Add no-op inline functions for the !(CONFIG_SCHED_MC || CONFIG_SCHED_SMT)
case.
This will eliminate all the #ifdef jungle in find_busiest_group() and the
other helper functions.
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Balbir Singh" <balbir@in.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: "Dhaval Giani" <dhaval@linux.vnet.ibm.com>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
Cc: "Vaidyanathan Srinivasan" <svaidy@linux.vnet.ibm.com>
LKML-Reference: <20090325091422.13992.73616.stgit@sofia.in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup, micro-optimization
We don't need to perform power_savings balance if either the
cpu is NOT_IDLE or if the sched_domain doesn't contain the
SD_POWERSAVINGS_BALANCE flag set.
Currently, we check for these conditions multiple number of
times, even though these variables don't change over the scope
of find_busiest_group().
Check once, and store the value in the already exiting
"power_savings_balance" variable.
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Balbir Singh" <balbir@in.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: "Dhaval Giani" <dhaval@linux.vnet.ibm.com>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
Cc: "Vaidyanathan Srinivasan" <svaidy@linux.vnet.ibm.com>
LKML-Reference: <20090325091417.13992.2657.stgit@sofia.in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Move all the imbalance calculation out of find_busiest_group()
through this helper function.
With this change, the structure of find_busiest_group() will be
as follows:
- update_sched_domain_statistics.
- check if imbalance exits.
- update imbalance and return busiest.
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Balbir Singh" <balbir@in.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: "Dhaval Giani" <dhaval@linux.vnet.ibm.com>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
Cc: "Vaidyanathan Srinivasan" <svaidy@linux.vnet.ibm.com>
LKML-Reference: <20090325091411.13992.43293.stgit@sofia.in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
We have two places in find_busiest_group() where we need to calculate
the minor imbalance before returning the busiest group. Encapsulate
this functionality into a seperate helper function.
Credit: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Balbir Singh" <balbir@in.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: "Dhaval Giani" <dhaval@linux.vnet.ibm.com>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
LKML-Reference: <20090325091406.13992.54316.stgit@sofia.in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
Create a helper function named update_sd_lb_stats() to update the
various sched_domain related statistics in find_busiest_group().
With this we would have moved all the statistics computation out of
find_busiest_group().
Credit: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Balbir Singh" <balbir@in.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: "Dhaval Giani" <dhaval@linux.vnet.ibm.com>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
LKML-Reference: <20090325091401.13992.88737.stgit@sofia.in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
Currently we use a lot of local variables in find_busiest_group()
to capture the various statistics related to the sched_domain.
Group them together into a single data structure.
This will help us to offload the job of updating the sched_domain
statistics to a helper function.
Credit: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Balbir Singh" <balbir@in.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: "Dhaval Giani" <dhaval@linux.vnet.ibm.com>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
LKML-Reference: <20090325091356.13992.25970.stgit@sofia.in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
Create a helper function named update_sg_lb_stats() which
can be invoked to calculate the individual group's statistics
in find_busiest_group().
This reduces the lenght of find_busiest_group() considerably.
Credit: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Aked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Balbir Singh" <balbir@in.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: "Dhaval Giani" <dhaval@linux.vnet.ibm.com>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
LKML-Reference: <20090325091351.13992.43461.stgit@sofia.in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
Currently a whole bunch of variables are used to store the
various statistics pertaining to the groups we iterate over
in find_busiest_group().
Group them together in a single data structure and add
appropriate comments.
This will be useful later on when we create helper functions
to calculate the sched_group statistics.
Credit: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Balbir Singh" <balbir@in.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: "Dhaval Giani" <dhaval@linux.vnet.ibm.com>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
LKML-Reference: <20090325091345.13992.20099.stgit@sofia.in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
Some indentations in find_busiest_group() can minimized by using
early exits with the help of gotos. This improves readability in
a couple of cases.
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Balbir Singh" <balbir@in.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: "Dhaval Giani" <dhaval@linux.vnet.ibm.com>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
Cc: "Vaidyanathan Srinivasan" <svaidy@linux.vnet.ibm.com>
LKML-Reference: <20090325091340.13992.45062.stgit@sofia.in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
Currently the load idx calculation code is in find_busiest_group().
Move that to a static inline helper function.
Similary, to find the first cpu of a sched_group we use
cpumask_first(sched_group_cpus(group))
Use a helper to that. It improves readability in some cases.
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Balbir Singh" <balbir@in.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: "Dhaval Giani" <dhaval@linux.vnet.ibm.com>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
Cc: "Vaidyanathan Srinivasan" <svaidy@linux.vnet.ibm.com>
LKML-Reference: <20090325091335.13992.55424.stgit@sofia.in.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup, new schedstat ABI
Since they are used on in statistics and are always set to zero, the
following fields from struct rq have been removed: yld_exp_empty,
yld_act_empty and yld_both_empty.
Both Sched Debug and SCHEDSTAT_VERSION versions has also been
incremented since ABIs have been changed.
The schedtop tool has been updated to properly handle new version of
schedstat:
http://rt.wiki.kernel.org/index.php/Schedtop_utility
Signed-off-by: Luis Henriques <henrix@sapo.pt>
Acked-by: Gregory Haskins <ghaskins@novell.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <20090324221002.GA10061@hades.domain.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix circular locking
Steven reports a circular locking from alloc_cpumask_var doing
a wakeup. We get rid of this using the tried-and-true technique
of using a per-cpu cpumask_var_t rather than doing an alloc
every time.
Simpler and more robust than a rare, implicit allocation within
an atomic codepath.
Reported-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
LKML-Reference: <alpine.DEB.2.00.0903181729360.31583@gandalf.stny.rr.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
There were 3 invocations of task_hot() in can_migrate_task().
Replace these 3 invocations by only one invocation, cached in
a local variable.
Signed-off-by: Luis Henriques <henrix@sapo.pt>
LKML-Reference: <20090316195902.GA6197@hades.domain.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Fixed typos in function documentation.
Signed-off-by: Luis Henriques <henrix@sapo.pt>
LKML-Reference: <20090316195809.GA6073@hades.domain.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
This is presumably what those definitions are for, and while all archs
define cpu_core_map/cpu_sibling map, that's changing (eg. x86 wants to
change it to a pointer).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Impact: micro-optimization
In cpuacct_charge(), task_ca() will never return NULL, so change
for(...) to do { } while(...) to save one NULL check.
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Menage <menage@google.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: Bharata B Rao <bharata@linux.vnet.ibm.com>
LKML-Reference: <49B863F5.2060400@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: more precise avg_overlap metric - better load-balancing
avg_overlap is used to measure the runtime overlap of the waker and
wakee.
However, when a process changes behaviour, eg a pipe becomes
un-congested and we don't need to go to sleep after a wakeup
for a while, the avg_overlap value grows stale.
When running we use the avg runtime between preemption as a
measure for avg_overlap since the amount of runtime can be
correlated to cache footprint.
The longer we run, the less likely we'll be wanting to be
migrated to another CPU.
Signed-off-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1236709131.25234.576.camel@laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: micro-optimization
We can avoid the sched domain walk on try_to_wake_up() when we know
there are no groups.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1236603381.8389.455.camel@laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
Use test_tsk_need_resched(), set_tsk_need_resched(), need_resched()
instead of using TIF_NEED_RESCHED.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <49B10BA4.9070209@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix function graph trace hang / drop pointless softirq on UP
While debugging a function graph trace hang on an old PII, I saw
that it consumed most of its time on the timer interrupt. And
the domain rebalancing softirq was the most concerned.
The timer interrupt calls trigger_load_balance() which will
decide if it is worth to schedule a rebalancing softirq.
In case of builtin UP kernel, no problem arises because there is
no domain question.
In case of builtin SMP kernel running on an SMP box, still no
problem, the softirq will be raised each time we reach the
next_balance time.
In case of builtin SMP kernel running on a UP box (most distros
provide default SMP kernels, whatever the box you have), then
the CPU is attached to the NULL sched domain. So a kind of
unexpected behaviour happen:
trigger_load_balance() -> raises the rebalancing softirq later
on softirq: run_rebalance_domains() -> rebalance_domains() where
the for_each_domain(cpu, sd) is not taken because of the NULL
domain we are attached at. Which means rq->next_balance is never
updated. So on the next timer tick, we will enter
trigger_load_balance() which will always reschedule() the
rebalacing softirq:
if (time_after_eq(jiffies, rq->next_balance))
raise_softirq(SCHED_SOFTIRQ);
So for each tick, we process this pointless softirq.
This patch fixes it by checking if we are attached to the null
domain before raising the softirq, another possible fix would be
to set the maximal possible JIFFIES value to rq->next_balance if
we are attached to the NULL domain.
v2: build fix on UP
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <49af242d.1c07d00a.32d5.ffffc019@mx.google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: micro-optimization
Parameter "prev" is not used really.
Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix hung task with certain (non-default) rt-limit settings
Corey Hickey reported that on using setuid to change the uid of a
rt process, the process would be unkillable and not be running.
This is because there was no rt runtime for that user group. Add
in a check to see if a user can attach an rt task to its task group.
On failure, return EINVAL, which is also returned in
CONFIG_CGROUP_SCHED.
Reported-by: Corey Hickey <bugfood-ml@fatooh.org>
Signed-off-by: Dhaval Giani <dhaval@linux.vnet.ibm.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix incorrect condition check
No need to start rt bandwidth timer when rt bandwidth is disabled.
If this timer starts, it may stop at sched_rt_period_timer() on the first time.
Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
cpuacct_charge() is in fast-path, and checking of !cpuacct_susys.active
always returns false after cpuacct has been initialized at system boot.
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Menage <menage@google.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Oleg noticed that we don't strictly need CSD_FLAG_WAIT, rework
the code so that we can use CSD_FLAG_LOCK for both purposes.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
There are two allocated per-cpu accessor macros with almost identical
spelling. The original and far more popular is per_cpu_ptr (44
files), so change over the other 4 files.
tj: kill percpu_ptr() and update UP too
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: mingo@redhat.com
Cc: lenb@kernel.org
Cc: cpufreq@vger.kernel.org
Signed-off-by: Tejun Heo <tj@kernel.org>
#define TASK_NICE(p) PRIO_TO_NICE((p)->static_prio)
So it's better to use TASK_NICE here.
Signed-off-by: WANG Cong <wangcong@zeuux.org>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: struct rq size optimization
The idle_at_tick in struct rq is only used in SMP settings
and it does not make sense to have this in the rq in an UP setup.
Signed-off-by: Henrik Austad <henrik@austad.us>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Intel reported a 10% regression (mysql+sysbench) on a 16-way machine
with these patches:
1596e29: sched: symmetric sync vs avg_overlap
d942fb6: sched: fix sync wakeups
Revert them.
Reported-by: "Zhang, Yanmin" <yanmin_zhang@linux.intel.com>
Bisected-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
With exclusive waiters, every process woken up through the wait queue must
ensure that the next waiter down the line is woken when it has finished.
Interruptible waiters don't do that when aborting due to a signal. And if
an aborting waiter is concurrently woken up through the waitqueue, noone
will ever wake up the next waiter.
This has been observed with __wait_on_bit_lock() used by
lock_page_killable(): the first contender on the queue was aborting when
the actual lock holder woke it up concurrently. The aborted contender
didn't acquire the lock and therefor never did an unlock followed by
waking up the next waiter.
Add abort_exclusive_wait() which removes the process' wait descriptor from
the waitqueue, iff still queued, or wakes up the next waiter otherwise.
It does so under the waitqueue lock. Racing with a wake up means the
aborting process is either already woken (removed from the queue) and will
wake up the next waiter, or it will remove itself from the queue and the
concurrent wake up will apply to the next waiter after it.
Use abort_exclusive_wait() in __wait_event_interruptible_exclusive() and
__wait_on_bit_lock() when they were interrupted by other means than a wake
up through the queue.
[akpm@linux-foundation.org: coding-style fixes]
Reported-by: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Mentored-by: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Chuck Lever <cel@citi.umich.edu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: <stable@kernel.org> ["after some testing"]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Christian Borntraeger reports:
> After a logical cpu offline, even on a complete idle system, there
> is one cpu with full ticks. It turns out that nohz.cpu_mask has the
> the offlined cpu still set.
>
> In select_nohz_load_balancer() we check if the system is completely
> idle to turn of load balancing. We compare cpu_online_map with
> nohz.cpu_mask. Since cpu_online_map is updated on cpu unplug,
> but nohz.cpu_mask is not, the check fails and the scheduler believes
> that we need an "idle load balancer" even on a fully idle system.
> Since the ilb cpu does not deactivate the timer tick this breaks NOHZ.
Fix the select_nohz_load_balancer() to not set the nohz.cpu_mask
while a cpu is going offline.
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Tested-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Reinstate the weakening of the sync hint if set. This yields a more
symmetric usage of avg_overlap.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Pawel Dziekonski reported that the openssl benchmark and his
quantum chemistry application both show slowdowns due to the
scheduler under-parallelizing execution.
The reason are pipe wakeups still doing 'sync' wakeups which
overrides the normal buddy wakeup logic - even if waker and
wakee are loosely coupled.
Fix an inversion of logic in the buddy wakeup code.
Reported-by: Pawel Dziekonski <dzieko@gmail.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: fix to preempt trace triggering lockdep check_flag failure
In local_bh_disable, the use of add_preempt_count causes the
preempt tracer to start recording the time preemption is off.
But because it already modified the preempt_count to show
softirqs disabled, and before it called the lockdep code to
handle this, it causes a state that lockdep can not handle.
The preempt tracer will reset the ring buffer on start of a trace,
and the ring buffer reset code does a spin_lock_irqsave. This
calls into lockdep and lockdep will fail when it detects the
invalid state of having softirqs disabled but the internal
current->softirqs_enabled is still set.
The fix is to manually add the SOFTIRQ_OFFSET to preempt count
and call the preempt tracer code outside the lockdep critical
area.
Thanks to Peter Zijlstra for suggesting this solution.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Increase the SCHED_IDLE weight from 2 to 3, this gives much more stable
vruntime numbers.
time advanced in 100ms:
weight=2
64765.988352
67012.881408
88501.412352
weight=3
35496.181411
34130.971298
35497.411573
Signed-off-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: make rt-limit tunables work again
Mark Glines reported:
> I've got an issue on x86-64 where I can't configure the system to allow
> RT tasks for a non-root user.
>
> In 2.6.26.5, I was able to do the following to set things up nicely:
> echo 450000 >/sys/kernel/uids/0/cpu_rt_runtime
> echo 450000 >/sys/kernel/uids/1000/cpu_rt_runtime
>
> Seems like every value I try to echo into the /sys files returns EINVAL.
For UID grouping we initialize the root group with infinite bandwidth
which by default is actually more than the global limit, therefore the
bandwidth check always fails.
Because the root group is a phantom group (for UID grouping) we cannot
runtime adjust it, therefore we let it reflect the global bandwidth
settings.
Reported-by: Mark Glines <mark@glines.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Introduce a new avg_wakeup statistic.
avg_wakeup is a measure of how frequently a task wakes up other tasks, it
represents the average time between wakeups, with a limit of avg_runtime
for when it doesn't wake up anybody.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Change mutex contention behaviour such that it will sometimes busy wait on
acquisition - moving its behaviour closer to that of spinlocks.
This concept got ported to mainline from the -rt tree, where it was originally
implemented for rtmutexes by Steven Rostedt, based on work by Gregory Haskins.
Testing with Ingo's test-mutex application (http://lkml.org/lkml/2006/1/8/50)
gave a 345% boost for VFS scalability on my testbox:
# ./test-mutex-shm V 16 10 | grep "^avg ops"
avg ops/sec: 296604
# ./test-mutex-shm V 16 10 | grep "^avg ops"
avg ops/sec: 85870
The key criteria for the busy wait is that the lock owner has to be running on
a (different) cpu. The idea is that as long as the owner is running, there is a
fair chance it'll release the lock soon, and thus we'll be better off spinning
instead of blocking/scheduling.
Since regular mutexes (as opposed to rtmutexes) do not atomically track the
owner, we add the owner in a non-atomic fashion and deal with the races in
the slowpath.
Furthermore, to ease the testing of the performance impact of this new code,
there is means to disable this behaviour runtime (without having to reboot
the system), when scheduler debugging is enabled (CONFIG_SCHED_DEBUG=y),
by issuing the following command:
# echo NO_OWNER_SPIN > /debug/sched_features
This command re-enables spinning again (this is also the default):
# echo OWNER_SPIN > /debug/sched_features
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The problem is that dropping the spinlock right before schedule is a voluntary
preemption point and can cause a schedule, right after which we schedule again.
Fix this inefficiency by keeping preemption disabled until we schedule, do this
by explicity disabling preemption and providing a schedule() variant that
assumes preemption is already disabled.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Davide Libenzi