Fix the insertion of cfs_rq in rq->leaf_cfs_rq_list to ensure that a
child will always be called before its parent.
The hierarchical order in shares update list has been introduced by
commit:
67e86250f8 ("sched: Introduce hierarchal order on shares update list")
With the current implementation a child can be still put after its
parent.
Lets take the example of:
root
\
b
/\
c d*
|
e*
with root -> b -> c already enqueued but not d -> e so the
leaf_cfs_rq_list looks like: head -> c -> b -> root -> tail
The branch d -> e will be added the first time that they are enqueued,
starting with e then d.
When e is added, its parents is not already on the list so e is put at
the tail : head -> c -> b -> root -> e -> tail
Then, d is added at the head because its parent is already on the
list: head -> d -> c -> b -> root -> e -> tail
e is not placed at the right position and will be called the last
whereas it should be called at the beginning.
Because it follows the bottom-up enqueue sequence, we are sure that we
will finished to add either a cfs_rq without parent or a cfs_rq with a
parent that is already on the list. We can use this event to detect
when we have finished to add a new branch. For the others, whose
parents are not already added, we have to ensure that they will be
added after their children that have just been inserted the steps
before, and after any potential parents that are already in the list.
The easiest way is to put the cfs_rq just after the last inserted one
and to keep track of it untl the branch is fully added.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Morten.Rasmussen@arm.com
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: bsegall@google.com
Cc: kernellwp@gmail.com
Cc: pjt@google.com
Cc: yuyang.du@intel.com
Link: http://lkml.kernel.org/r/1478598827-32372-3-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
struct sched_group_capacity currently represents the compute capacity
sum of all CPUs in the sched_group.
Unless it is divided by the group_weight to get the average capacity
per CPU, it hides differences in CPU capacity for mixed capacity systems
(e.g. high RT/IRQ utilization or ARM big.LITTLE).
But even the average may not be sufficient if the group covers CPUs of
different capacities.
Instead, by extending struct sched_group_capacity to indicate min per-CPU
capacity in the group a suitable group for a given task utilization can
more easily be found such that CPUs with reduced capacity can be avoided
for tasks with high utilization (not implemented by this patch).
Signed-off-by: Morten Rasmussen <morten.rasmussen@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: dietmar.eggemann@arm.com
Cc: freedom.tan@mediatek.com
Cc: keita.kobayashi.ym@renesas.com
Cc: mgalbraith@suse.de
Cc: sgurrappadi@nvidia.com
Cc: vincent.guittot@linaro.org
Cc: yuyang.du@intel.com
Link: http://lkml.kernel.org/r/1476452472-24740-4-git-send-email-morten.rasmussen@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When CONFIG_THREAD_INFO_IN_TASK=y, it is possible that an exited thread
remains in the task list after its stack pointer was already set to NULL.
Therefore, thread_saved_pc() and stack_not_used() in sched_show_task()
will trigger NULL pointer dereference if an attempt to dump such thread's
traces (e.g. SysRq-t, khungtaskd) is made.
Since show_stack() in sched_show_task() calls try_get_task_stack() and
sched_show_task() is called from interrupt context, calling
try_get_task_stack() from sched_show_task() will be safe as well.
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Acked-by: Andy Lutomirski <luto@kernel.org>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: bp@alien8.de
Cc: brgerst@gmail.com
Cc: jann@thejh.net
Cc: keescook@chromium.org
Cc: linux-api@vger.kernel.org
Cc: tycho.andersen@canonical.com
Link: http://lkml.kernel.org/r/201611021950.FEJ34368.HFFJOOMLtQOVSF@I-love.SAKURA.ne.jp
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The per-zone waitqueues exist because of a scalability issue with the
page waitqueues on some NUMA machines, but it turns out that they hurt
normal loads, and now with the vmalloced stacks they also end up
breaking gfs2 that uses a bit_wait on a stack object:
wait_on_bit(&gh->gh_iflags, HIF_WAIT, TASK_UNINTERRUPTIBLE)
where 'gh' can be a reference to the local variable 'mount_gh' on the
stack of fill_super().
The reason the per-zone hash table breaks for this case is that there is
no "zone" for virtual allocations, and trying to look up the physical
page to get at it will fail (with a BUG_ON()).
It turns out that I actually complained to the mm people about the
per-zone hash table for another reason just a month ago: the zone lookup
also hurts the regular use of "unlock_page()" a lot, because the zone
lookup ends up forcing several unnecessary cache misses and generates
horrible code.
As part of that earlier discussion, we had a much better solution for
the NUMA scalability issue - by just making the page lock have a
separate contention bit, the waitqueue doesn't even have to be looked at
for the normal case.
Peter Zijlstra already has a patch for that, but let's see if anybody
even notices. In the meantime, let's fix the actual gfs2 breakage by
simplifying the bitlock waitqueues and removing the per-zone issue.
Reported-by: Andreas Gruenbacher <agruenba@redhat.com>
Tested-by: Bob Peterson <rpeterso@redhat.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Steven Whitehouse <swhiteho@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The current mutex implementation has an atomic lock word and a
non-atomic owner field.
This disparity leads to a number of issues with the current mutex code
as it means that we can have a locked mutex without an explicit owner
(because the owner field has not been set, or already cleared).
This leads to a number of weird corner cases, esp. between the
optimistic spinning and debug code. Where the optimistic spinning
code needs the owner field updated inside the lock region, the debug
code is more relaxed because the whole lock is serialized by the
wait_lock.
Also, the spinning code itself has a few corner cases where we need to
deal with a held lock without an owner field.
Furthermore, it becomes even more of a problem when trying to fix
starvation cases in the current code. We end up stacking special case
on special case.
To solve this rework the basic mutex implementation to be a single
atomic word that contains the owner and uses the low bits for extra
state.
This matches how PI futexes and rt_mutex already work. By having the
owner an integral part of the lock state a lot of the problems
dissapear and we get a better option to deal with starvation cases,
direct owner handoff.
Changing the basic mutex does however invalidate all the arch specific
mutex code; this patch leaves that unused in-place, a later patch will
remove that.
Tested-by: Jason Low <jason.low2@hpe.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Will Deacon <will.deacon@arm.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
There were a few questions wrt. how sleep-wakeup works. Try and explain
it more.
Requested-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull low-level x86 updates from Ingo Molnar:
"In this cycle this topic tree has become one of those 'super topics'
that accumulated a lot of changes:
- Add CONFIG_VMAP_STACK=y support to the core kernel and enable it on
x86 - preceded by an array of changes. v4.8 saw preparatory changes
in this area already - this is the rest of the work. Includes the
thread stack caching performance optimization. (Andy Lutomirski)
- switch_to() cleanups and all around enhancements. (Brian Gerst)
- A large number of dumpstack infrastructure enhancements and an
unwinder abstraction. The secret long term plan is safe(r) live
patching plus maybe another attempt at debuginfo based unwinding -
but all these current bits are standalone enhancements in a frame
pointer based debug environment as well. (Josh Poimboeuf)
- More __ro_after_init and const annotations. (Kees Cook)
- Enable KASLR for the vmemmap memory region. (Thomas Garnier)"
[ The virtually mapped stack changes are pretty fundamental, and not
x86-specific per se, even if they are only used on x86 right now. ]
* 'x86-asm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (70 commits)
x86/asm: Get rid of __read_cr4_safe()
thread_info: Use unsigned long for flags
x86/alternatives: Add stack frame dependency to alternative_call_2()
x86/dumpstack: Fix show_stack() task pointer regression
x86/dumpstack: Remove dump_trace() and related callbacks
x86/dumpstack: Convert show_trace_log_lvl() to use the new unwinder
oprofile/x86: Convert x86_backtrace() to use the new unwinder
x86/stacktrace: Convert save_stack_trace_*() to use the new unwinder
perf/x86: Convert perf_callchain_kernel() to use the new unwinder
x86/unwind: Add new unwind interface and implementations
x86/dumpstack: Remove NULL task pointer convention
fork: Optimize task creation by caching two thread stacks per CPU if CONFIG_VMAP_STACK=y
sched/core: Free the stack early if CONFIG_THREAD_INFO_IN_TASK
lib/syscall: Pin the task stack in collect_syscall()
x86/process: Pin the target stack in get_wchan()
x86/dumpstack: Pin the target stack when dumping it
kthread: Pin the stack via try_get_task_stack()/put_task_stack() in to_live_kthread() function
sched/core: Add try_get_task_stack() and put_task_stack()
x86/entry/64: Fix a minor comment rebase error
iommu/amd: Don't put completion-wait semaphore on stack
...
Almost all scheduler functions update state with the following
pattern:
if (queued)
dequeue_task(rq, p, DEQUEUE_SAVE);
if (running)
put_prev_task(rq, p);
/* update state */
if (queued)
enqueue_task(rq, p, ENQUEUE_RESTORE);
if (running)
set_curr_task(rq, p);
set_user_nice() however misses the running part, cure this.
This was found by asserting we never enqueue 'current'.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Now that the ia64 only set_curr_task() symbol is gone, provide a
helper just like put_prev_task().
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Rename the ia64 only set_curr_task() function to free up the name.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tony Luck <tony.luck@intel.com>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When a task switches to fair scheduling class, the period between now
and the last update of its utilization is accounted as running time
whatever happened during this period. This incorrect accounting applies
to the task and also to the task group branch.
When changing the property of a running task like its list of allowed
CPUs or its scheduling class, we follow the sequence:
- dequeue task
- put task
- change the property
- set task as current task
- enqueue task
The end of the sequence doesn't follow the normal sequence (as per
__schedule()) which is:
- enqueue a task
- then set the task as current task.
This incorrectordering is the root cause of incorrect utilization accounting.
Update the sequence to follow the right one:
- dequeue task
- put task
- change the property
- enqueue task
- set task as current task
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Morten.Rasmussen@arm.com
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: bsegall@google.com
Cc: dietmar.eggemann@arm.com
Cc: linaro-kernel@lists.linaro.org
Cc: pjt@google.com
Cc: yuyang.du@intel.com
Link: http://lkml.kernel.org/r/1473666472-13749-8-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
select_idle_siblings() is a known pain point for a number of
workloads; it either does too much or not enough and sometimes just
does plain wrong.
This rewrite attempts to address a number of issues (but sadly not
all).
The current code does an unconditional sched_domain iteration; with
the intent of finding an idle core (on SMT hardware). The problems
which this patch tries to address are:
- its pointless to look for idle cores if the machine is real busy;
at which point you're just wasting cycles.
- it's behaviour is inconsistent between SMT and !SMT hardware in
that !SMT hardware ends up doing a scan for any idle CPU in the LLC
domain, while SMT hardware does a scan for idle cores and if that
fails, falls back to a scan for idle threads on the 'target' core.
The new code replaces the sched_domain scan with 3 explicit scans:
1) search for an idle core in the LLC
2) search for an idle CPU in the LLC
3) search for an idle thread in the 'target' core
where 1 and 3 are conditional on SMT support and 1 and 2 have runtime
heuristics to skip the step.
Step 1) is conditional on sd_llc_shared->has_idle_cores; when a cpu
goes idle and sd_llc_shared->has_idle_cores is false, we scan all SMT
siblings of the CPU going idle. Similarly, we clear
sd_llc_shared->has_idle_cores when we fail to find an idle core.
Step 2) tracks the average cost of the scan and compares this to the
average idle time guestimate for the CPU doing the wakeup. There is a
significant fudge factor involved to deal with the variability of the
averages. Esp. hackbench was sensitive to this.
Step 3) is unconditional; we assume (also per step 1) that scanning
all SMT siblings in a core is 'cheap'.
With this; SMT systems gain step 2, which cures a few benchmarks --
notably one from Facebook.
One 'feature' of the sched_domain iteration, which we preserve in the
new code, is that it would start scanning from the 'target' CPU,
instead of scanning the cpumask in cpu id order. This avoids multiple
CPUs in the LLC scanning for idle to gang up and find the same CPU
quite as much. The down side is that tasks can end up hopping across
the LLC for no apparent reason.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Move the nr_busy_cpus thing from its hacky sd->parent->groups->sgc
location into the much more natural sched_domain_shared location.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Since struct sched_domain is strictly per cpu; introduce a structure
that is shared between all 'identical' sched_domains.
Limit to SD_SHARE_PKG_RESOURCES domains for now, as we'll only use it
for shared cache state; if another use comes up later we can easily
relax this.
While the sched_group's are normally shared between CPUs, these are
not natural to use when we need some shared state on a domain level --
since that would require the domain to have a parent, which is not a
given.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
There is no point in doing a call_rcu() for each domain, only do a
callback for the root sched domain and clean up the entire set in one
go.
Also make the entire call chain be called destroy_sched_domain*() to
remove confusion with the free_sched_domains() call, which does an
entirely different thing.
Both cpu_attach_domain() callers of destroy_sched_domain() can live
without the call_rcu() because at those points the sched_domain hasn't
been published yet.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Small cleanup; nothing uses the @cpu argument so make it go away.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Current code can call set_cpu_sibling_map() and invoke sched_set_topology()
more than once (e.g. on CPU hot plug). When this happens after
sched_init_smp() has been called, we lose the NUMA topology extension to
sched_domain_topology in sched_init_numa(). This results in incorrect
topology when the sched domain is rebuilt.
This patch fixes the bug and issues warning if we call sched_set_topology()
after sched_init_smp().
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: bp@suse.de
Cc: jolsa@redhat.com
Cc: rjw@rjwysocki.net
Link: http://lkml.kernel.org/r/1474485552-141429-2-git-send-email-srinivas.pandruvada@linux.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
On fully preemptible kernels _cond_resched() is pointless, so avoid
emitting any code for it.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mikulas Patocka <mpatocka@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Oleg noted that by making do_exit() use __schedule() for the TASK_DEAD
context switch, we can avoid the TASK_DEAD special case currently in
__schedule() because that avoids the extra preempt_disable() from
schedule().
In order to facilitate this, create a do_task_dead() helper which we
place in the scheduler code, such that it can access __schedule().
Also add some __noreturn annotations to the functions, there's no
coming back from do_exit().
Suggested-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Cheng Chao <cs.os.kernel@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: akpm@linux-foundation.org
Cc: chris@chris-wilson.co.uk
Cc: tj@kernel.org
Link: http://lkml.kernel.org/r/20160913163729.GB5012@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In case @cpu == smp_proccessor_id(), we can avoid a sleep+wakeup
cycle by doing a preemption.
Callers such as sched_exec() can benefit from this change.
Signed-off-by: Cheng Chao <cs.os.kernel@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: akpm@linux-foundation.org
Cc: chris@chris-wilson.co.uk
Cc: tj@kernel.org
Link: http://lkml.kernel.org/r/1473818510-6779-1-git-send-email-cs.os.kernel@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
We currently keep every task's stack around until the task_struct
itself is freed. This means that we keep the stack allocation alive
for longer than necessary and that, under load, we free stacks in
big batches whenever RCU drops the last task reference. Neither of
these is good for reuse of cache-hot memory, and freeing in batches
prevents us from usefully caching small numbers of vmalloced stacks.
On architectures that have thread_info on the stack, we can't easily
change this, but on architectures that set THREAD_INFO_IN_TASK, we
can free it as soon as the task is dead.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Jann Horn <jann@thejh.net>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/08ca06cde00ebed0046c5d26cbbf3fbb7ef5b812.1474003868.git.luto@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull RCU changes from Paul E. McKenney:
- Expedited grace-period changes, most notably avoiding having
user threads drive expedited grace periods, using a workqueue
instead.
- Miscellaneous fixes, including a performance fix for lists
that was sent with the lists modifications (second URL below).
- CPU hotplug updates, most notably providing exact CPU-online
tracking for RCU. This will in turn allow removal of the
checks supporting RCU's prior heuristic that was based on the
assumption that CPUs would take no longer than one jiffy to
come online.
- Torture-test updates.
- Documentation updates.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The schedstat_*() macros are inconsistent: most of them take a pointer
and a field which the macro combines, whereas schedstat_set() takes the
already combined ptr->field.
The already combined ptr->field argument is actually more intuitive and
easier to use, and there's no reason to require the user to split the
variable up, so convert the macros to use the combined argument.
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matt Fleming <matt@codeblueprint.co.uk>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/54953ca25bb579f3a5946432dee409b0e05222c6.1466184592.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
init_task's preempt_notifiers is initialized twice:
1) sched_init()
-> INIT_HLIST_HEAD(&init_task.preempt_notifiers)
2) sched_init()
-> init_idle(current,) <--- current task is init_task at this time
-> __sched_fork(,current)
-> INIT_HLIST_HEAD(&p->preempt_notifiers)
I think the first one is unnecessary, so remove it.
Signed-off-by: seokhoon.yoon <iamyooon@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1471339568-5790-1-git-send-email-iamyooon@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The origin of the issue I've seen is related to
a missing memory barrier between check for task->state and
the check for task->on_rq.
The task being woken up is already awake from a schedule()
and is doing the following:
do {
schedule()
set_current_state(TASK_(UN)INTERRUPTIBLE);
} while (!cond);
The waker, actually gets stuck doing the following in
try_to_wake_up():
while (p->on_cpu)
cpu_relax();
Analysis:
The instance I've seen involves the following race:
CPU1 CPU2
while () {
if (cond)
break;
do {
schedule();
set_current_state(TASK_UN..)
} while (!cond);
wakeup_routine()
spin_lock_irqsave(wait_lock)
raw_spin_lock_irqsave(wait_lock) wake_up_process()
} try_to_wake_up()
set_current_state(TASK_RUNNING); ..
list_del(&waiter.list);
CPU2 wakes up CPU1, but before it can get the wait_lock and set
current state to TASK_RUNNING the following occurs:
CPU3
wakeup_routine()
raw_spin_lock_irqsave(wait_lock)
if (!list_empty)
wake_up_process()
try_to_wake_up()
raw_spin_lock_irqsave(p->pi_lock)
..
if (p->on_rq && ttwu_wakeup())
..
while (p->on_cpu)
cpu_relax()
..
CPU3 tries to wake up the task on CPU1 again since it finds
it on the wait_queue, CPU1 is spinning on wait_lock, but immediately
after CPU2, CPU3 got it.
CPU3 checks the state of p on CPU1, it is TASK_UNINTERRUPTIBLE and
the task is spinning on the wait_lock. Interestingly since p->on_rq
is checked under pi_lock, I've noticed that try_to_wake_up() finds
p->on_rq to be 0. This was the most confusing bit of the analysis,
but p->on_rq is changed under runqueue lock, rq_lock, the p->on_rq
check is not reliable without this fix IMHO. The race is visible
(based on the analysis) only when ttwu_queue() does a remote wakeup
via ttwu_queue_remote. In which case the p->on_rq change is not
done uder the pi_lock.
The result is that after a while the entire system locks up on
the raw_spin_irqlock_save(wait_lock) and the holder spins infintely
Reproduction of the issue:
The issue can be reproduced after a long run on my system with 80
threads and having to tweak available memory to very low and running
memory stress-ng mmapfork test. It usually takes a long time to
reproduce. I am trying to work on a test case that can reproduce
the issue faster, but thats work in progress. I am still testing the
changes on my still in a loop and the tests seem OK thus far.
Big thanks to Benjamin and Nick for helping debug this as well.
Ben helped catch the missing barrier, Nick caught every missing
bit in my theory.
Signed-off-by: Balbir Singh <bsingharora@gmail.com>
[ Updated comment to clarify matching barriers. Many
architectures do not have a full barrier in switch_to()
so that cannot be relied upon. ]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Alexey Kardashevskiy <aik@ozlabs.ru>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Nicholas Piggin <nicholas.piggin@gmail.com>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/e02cce7b-d9ca-1ad0-7a61-ea97c7582b37@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Now that the x86 switch_to() uses the standard C calling convention,
the STACK_FRAME_NON_STANDARD() annotation is no longer needed.
Suggested-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Brian Gerst <brgerst@gmail.com>
Reviewed-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1471106302-10159-8-git-send-email-brgerst@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Both timers and hrtimers are maintained on the outgoing CPU until
CPU_DEAD time, at which point they are migrated to a surviving CPU. If a
mod_timer() executes between CPU_DYING and CPU_DEAD time, x86 systems
will splat in native_smp_send_reschedule() when attempting to wake up
the just-now-offlined CPU, as shown below from a NO_HZ_FULL kernel:
[ 7976.741556] WARNING: CPU: 0 PID: 661 at /home/paulmck/public_git/linux-rcu/arch/x86/kernel/smp.c:125 native_smp_send_reschedule+0x39/0x40
[ 7976.741595] Modules linked in:
[ 7976.741595] CPU: 0 PID: 661 Comm: rcu_torture_rea Not tainted 4.7.0-rc2+ #1
[ 7976.741595] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011
[ 7976.741595] 0000000000000000 ffff88000002fcc8 ffffffff8138ab2e 0000000000000000
[ 7976.741595] 0000000000000000 ffff88000002fd08 ffffffff8105cabc 0000007d1fd0ee18
[ 7976.741595] 0000000000000001 ffff88001fd16d40 ffff88001fd0ee00 ffff88001fd0ee00
[ 7976.741595] Call Trace:
[ 7976.741595] [<ffffffff8138ab2e>] dump_stack+0x67/0x99
[ 7976.741595] [<ffffffff8105cabc>] __warn+0xcc/0xf0
[ 7976.741595] [<ffffffff8105cb98>] warn_slowpath_null+0x18/0x20
[ 7976.741595] [<ffffffff8103cba9>] native_smp_send_reschedule+0x39/0x40
[ 7976.741595] [<ffffffff81089bc2>] wake_up_nohz_cpu+0x82/0x190
[ 7976.741595] [<ffffffff810d275a>] internal_add_timer+0x7a/0x80
[ 7976.741595] [<ffffffff810d3ee7>] mod_timer+0x187/0x2b0
[ 7976.741595] [<ffffffff810c89dd>] rcu_torture_reader+0x33d/0x380
[ 7976.741595] [<ffffffff810c66f0>] ? sched_torture_read_unlock+0x30/0x30
[ 7976.741595] [<ffffffff810c86a0>] ? rcu_bh_torture_read_lock+0x80/0x80
[ 7976.741595] [<ffffffff8108068f>] kthread+0xdf/0x100
[ 7976.741595] [<ffffffff819dd83f>] ret_from_fork+0x1f/0x40
[ 7976.741595] [<ffffffff810805b0>] ? kthread_create_on_node+0x200/0x200
However, in this case, the wakeup is redundant, because the timer
migration will reprogram timer hardware as needed. Note that the fact
that preemption is disabled does not avoid the splat, as the offline
operation has already passed both the synchronize_sched() and the
stop_machine() that would be blocked by disabled preemption.
This commit therefore modifies wake_up_nohz_cpu() to avoid attempting
to wake up offline CPUs. It also adds a comment stating that the
caller must tolerate lost wakeups when the target CPU is going offline,
and suggesting the CPU_DEAD notifier as a recovery mechanism.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
To be able to compare the capacity of the target CPU with the highest
available CPU capacity, store the maximum per-CPU capacity in the root
domain.
The max per-CPU capacity should be 1024 for all systems except SMT,
where the capacity is currently based on smt_gain and the number of
hardware threads and is <1024. If SMT can be brought to work with a
per-thread capacity of 1024, this patch can be dropped and replaced by a
hard-coded max capacity of 1024 (=SCHED_CAPACITY_SCALE).
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Morten Rasmussen <morten.rasmussen@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: freedom.tan@mediatek.com
Cc: keita.kobayashi.ym@renesas.com
Cc: mgalbraith@suse.de
Cc: sgurrappadi@nvidia.com
Cc: vincent.guittot@linaro.org
Cc: yuyang.du@intel.com
Link: http://lkml.kernel.org/r/26c69258-9947-f830-a53e-0c54e7750646@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
A domain with the SD_ASYM_CPUCAPACITY flag set indicate that
sched_groups at this level and below do not include CPUs of all
capacities available (e.g. group containing little-only or big-only CPUs
in big.LITTLE systems). It is therefore necessary to put in more effort
in finding an appropriate CPU at task wake-up by enabling balancing at
wake-up (SD_BALANCE_WAKE) on all lower (child) levels.
Signed-off-by: Morten Rasmussen <morten.rasmussen@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: dietmar.eggemann@arm.com
Cc: freedom.tan@mediatek.com
Cc: keita.kobayashi.ym@renesas.com
Cc: mgalbraith@suse.de
Cc: sgurrappadi@nvidia.com
Cc: vincent.guittot@linaro.org
Cc: yuyang.du@intel.com
Link: http://lkml.kernel.org/r/1469453670-2660-8-git-send-email-morten.rasmussen@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Add a topology flag to the sched_domain hierarchy indicating the lowest
domain level where the full range of CPU capacities is represented by
the domain members for asymmetric capacity topologies (e.g. ARM
big.LITTLE).
The flag is intended to indicate that extra care should be taken when
placing tasks on CPUs and this level spans all the different types of
CPUs found in the system (no need to look further up the domain
hierarchy). This information is currently only available through
iterating through the capacities of all the CPUs at parent levels in the
sched_domain hierarchy.
SD 2 [ 0 1 2 3] SD_ASYM_CPUCAPACITY
SD 1 [ 0 1] [ 2 3] !SD_ASYM_CPUCAPACITY
CPU: 0 1 2 3
capacity: 756 756 1024 1024
If the topology in the example above is duplicated to create an eight
CPU example with third sched_domain level on top (SD 3), this level
should not have the flag set (!SD_ASYM_CPUCAPACITY) as its two group
would both have all CPU capacities represented within them.
Signed-off-by: Morten Rasmussen <morten.rasmussen@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: dietmar.eggemann@arm.com
Cc: freedom.tan@mediatek.com
Cc: keita.kobayashi.ym@renesas.com
Cc: mgalbraith@suse.de
Cc: sgurrappadi@nvidia.com
Cc: vincent.guittot@linaro.org
Cc: yuyang.du@intel.com
Link: http://lkml.kernel.org/r/1469453670-2660-6-git-send-email-morten.rasmussen@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This message is currently really useless since it always prints a value
that comes from the printk() we just did, e.g.:
BUG: sleeping function called from invalid context at mm/slab.h:388
in_atomic(): 0, irqs_disabled(): 0, pid: 31996, name: trinity-c1
Preemption disabled at:[<ffffffff8119db33>] down_trylock+0x13/0x80
BUG: sleeping function called from invalid context at include/linux/freezer.h:56
in_atomic(): 0, irqs_disabled(): 0, pid: 31996, name: trinity-c1
Preemption disabled at:[<ffffffff811aaa37>] console_unlock+0x2f7/0x930
Here, both down_trylock() and console_unlock() is somewhere in the
printk() path.
We should save the value before calling printk() and use the saved value
instead. That immediately reveals the offending callsite:
BUG: sleeping function called from invalid context at mm/slab.h:388
in_atomic(): 0, irqs_disabled(): 0, pid: 14971, name: trinity-c2
Preemption disabled at:[<ffffffff819bcd46>] rhashtable_walk_start+0x46/0x150
Bug report:
http://marc.info/?l=linux-netdev&m=146925979821849&w=2
Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rusty Russel <rusty@rustcorp.com.au>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Add documentation for the cookie argument in try_to_wake_up_local().
This caused the following warning when building documentation:
kernel/sched/core.c:2088: warning: No description found for parameter 'cookie'
Signed-off-by: Luis de Bethencourt <luisbg@osg.samsung.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: akpm@linux-foundation.org
Fixes: e7904a28f5 ("ilocking/lockdep, sched/core: Implement a better lock pinning scheme")
Link: http://lkml.kernel.org/r/1468159226-17674-1-git-send-email-luisbg@osg.samsung.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Fix one minor typo in the comment: s/targer/target/.
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Link: http://lkml.kernel.org/r/1470378758-15066-1-git-send-email-leo.yan@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Commit:
6e998916df ("sched/cputime: Fix clock_nanosleep()/clock_gettime() inconsistency")
fixed a problem whereby clock_nanosleep() followed by clock_gettime() could
allow a task to wake early. It addressed the problem by calling the scheduling
classes update_curr() when the cputimer starts.
Said change induced a considerable performance regression on the syscalls
times() and clock_gettimes(CLOCK_PROCESS_CPUTIME_ID). There are some
debuggers and applications that monitor their own performance that
accidentally depend on the performance of these specific calls.
This patch mitigates the performace loss by prefetching data in the CPU
cache, as stalls due to cache misses appear to be where most time is spent
in our benchmarks.
Here are the performance gain of this patch over v4.7-rc7 on a Sandy Bridge
box with 32 logical cores and 2 NUMA nodes. The test is repeated with a
variable number of threads, from 2 to 4*num_cpus; the results are in
seconds and correspond to the average of 10 runs; the percentage gain is
computed with (before-after)/before so a positive value is an improvement
(it's faster). The improvement varies between a few percents for 5-20
threads and more than 10% for 2 or >20 threads.
pound_clock_gettime:
threads 4.7-rc7 patched 4.7-rc7
[num] [secs] [secs (percent)]
2 3.48 3.06 ( 11.83%)
5 3.33 3.25 ( 2.40%)
8 3.37 3.26 ( 3.30%)
12 3.32 3.37 ( -1.60%)
21 4.01 3.90 ( 2.74%)
30 3.63 3.36 ( 7.41%)
48 3.71 3.11 ( 16.27%)
79 3.75 3.16 ( 15.74%)
110 3.81 3.25 ( 14.80%)
128 3.88 3.31 ( 14.76%)
pound_times:
threads 4.7-rc7 patched 4.7-rc7
[num] [secs] [secs (percent)]
2 3.65 3.25 ( 11.03%)
5 3.45 3.17 ( 7.92%)
8 3.52 3.22 ( 8.69%)
12 3.29 3.36 ( -2.04%)
21 4.07 3.92 ( 3.78%)
30 3.87 3.40 ( 12.17%)
48 3.79 3.16 ( 16.61%)
79 3.88 3.28 ( 15.42%)
110 3.90 3.38 ( 13.35%)
128 4.00 3.38 ( 15.45%)
pound_clock_gettime and pound_clock_gettime are two benchmarks included in
the MMTests framework. They launch a given number of threads which
repeatedly call times() or clock_gettimes(). The results above can be
reproduced with cloning MMTests from github.com and running the "poundtime"
workload:
$ git clone https://github.com/gormanm/mmtests.git
$ cd mmtests
$ cp configs/config-global-dhp__workload_poundtime config
$ ./run-mmtests.sh --run-monitor $(uname -r)
The above will run "poundtime" measuring the kernel currently running on
the machine; Once a new kernel is installed and the machine rebooted,
running again
$ cd mmtests
$ ./run-mmtests.sh --run-monitor $(uname -r)
will produce results to compare with. A comparison table will be output
with:
$ cd mmtests/work/log
$ ../../compare-kernels.sh
the table will contain a lot of entries; grepping for "Amean" (as in
"arithmetic mean") will give the tables presented above. The source code
for the two benchmarks is reported at the end of this changelog for
clairity.
The cache misses addressed by this patch were found using a combination of
`perf top`, `perf record` and `perf annotate`. The incriminated lines were
found to be
struct sched_entity *curr = cfs_rq->curr;
and
delta_exec = now - curr->exec_start;
in the function update_curr() from kernel/sched/fair.c. This patch
prefetches the data from memory just before update_curr is called in the
interested execution path.
A comparison of the total number of cycles before and after the patch
follows; the data is obtained using `perf stat -r 10 -ddd <program>`
running over the same sequence of number of threads used above (a positive
gain is an improvement):
threads cycles before cycles after gain
2 19,699,563,964 +-1.19% 17,358,917,517 +-1.85% 11.88%
5 47,401,089,566 +-2.96% 45,103,730,829 +-0.97% 4.85%
8 80,923,501,004 +-3.01% 71,419,385,977 +-0.77% 11.74%
12 112,326,485,473 +-0.47% 110,371,524,403 +-0.47% 1.74%
21 193,455,574,299 +-0.72% 180,120,667,904 +-0.36% 6.89%
30 315,073,519,013 +-1.64% 271,222,225,950 +-1.29% 13.92%
48 321,969,515,332 +-1.48% 273,353,977,321 +-1.16% 15.10%
79 337,866,003,422 +-0.97% 289,462,481,538 +-1.05% 14.33%
110 338,712,691,920 +-0.78% 290,574,233,170 +-0.77% 14.21%
128 348,384,794,006 +-0.50% 292,691,648,206 +-0.66% 15.99%
A comparison of cache miss vs total cache loads ratios, before and after
the patch (again from the `perf stat -r 10 -ddd <program>` tables):
threads L1 misses/total*100 L1 misses/total*100 gain
before after
2 7.43 +-4.90% 7.36 +-4.70% 0.94%
5 13.09 +-4.74% 13.52 +-3.73% -3.28%
8 13.79 +-5.61% 12.90 +-3.27% 6.45%
12 11.57 +-2.44% 8.71 +-1.40% 24.72%
21 12.39 +-3.92% 9.97 +-1.84% 19.53%
30 13.91 +-2.53% 11.73 +-2.28% 15.67%
48 13.71 +-1.59% 12.32 +-1.97% 10.14%
79 14.44 +-0.66% 13.40 +-1.06% 7.20%
110 15.86 +-0.50% 14.46 +-0.59% 8.83%
128 16.51 +-0.32% 15.06 +-0.78% 8.78%
As a final note, the following shows the evolution of performance figures
in the "poundtime" benchmark and pinpoints commit 6e998916df
("sched/cputime: Fix clock_nanosleep()/clock_gettime() inconsistency") as a
major source of degradation, mostly unaddressed to this day (figures
expressed in seconds).
pound_clock_gettime:
threads parent of 6e998916df 4.7-rc7
6e998916df itself
2 2.23 3.68 ( -64.56%) 3.48 (-55.48%)
5 2.83 3.78 ( -33.42%) 3.33 (-17.43%)
8 2.84 4.31 ( -52.12%) 3.37 (-18.76%)
12 3.09 3.61 ( -16.74%) 3.32 ( -7.17%)
21 3.14 4.63 ( -47.36%) 4.01 (-27.71%)
30 3.28 5.75 ( -75.37%) 3.63 (-10.80%)
48 3.02 6.05 (-100.56%) 3.71 (-22.99%)
79 2.88 6.30 (-118.90%) 3.75 (-30.26%)
110 2.95 6.46 (-119.00%) 3.81 (-29.24%)
128 3.05 6.42 (-110.08%) 3.88 (-27.04%)
pound_times:
threads parent of 6e998916df 4.7-rc7
6e998916df itself
2 2.27 3.73 ( -64.71%) 3.65 (-61.14%)
5 2.78 3.77 ( -35.56%) 3.45 (-23.98%)
8 2.79 4.41 ( -57.71%) 3.52 (-26.05%)
12 3.02 3.56 ( -17.94%) 3.29 ( -9.08%)
21 3.10 4.61 ( -48.74%) 4.07 (-31.34%)
30 3.33 5.75 ( -72.53%) 3.87 (-16.01%)
48 2.96 6.06 (-105.04%) 3.79 (-28.10%)
79 2.88 6.24 (-116.83%) 3.88 (-34.81%)
110 2.98 6.37 (-114.08%) 3.90 (-31.12%)
128 3.10 6.35 (-104.61%) 4.00 (-28.87%)
The source code of the two benchmarks follows. To compile the two:
NR_THREADS=42
for FILE in pound_times pound_clock_gettime; do
gcc -lrt -O2 -lpthread -DNUM_THREADS=$NR_THREADS $FILE.c -o $FILE
done
==== BEGIN pound_times.c ====
struct tms start;
void *pound (void *threadid)
{
struct tms end;
int oldutime = 0;
int utime;
int i;
for (i = 0; i < 5000000 / NUM_THREADS; i++) {
times(&end);
utime = ((int)end.tms_utime - (int)start.tms_utime);
if (oldutime > utime) {
printf("utime decreased, was %d, now %d!\n", oldutime, utime);
}
oldutime = utime;
}
pthread_exit(NULL);
}
int main()
{
pthread_t th[NUM_THREADS];
long i;
times(&start);
for (i = 0; i < NUM_THREADS; i++) {
pthread_create (&th[i], NULL, pound, (void *)i);
}
pthread_exit(NULL);
return 0;
}
==== END pound_times.c ====
==== BEGIN pound_clock_gettime.c ====
void *pound (void *threadid)
{
struct timespec ts;
int rc, i;
unsigned long prev = 0, this = 0;
for (i = 0; i < 5000000 / NUM_THREADS; i++) {
rc = clock_gettime(CLOCK_PROCESS_CPUTIME_ID, &ts);
if (rc < 0)
perror("clock_gettime");
this = (ts.tv_sec * 1000000000) + ts.tv_nsec;
if (0 && this < prev)
printf("%lu ns timewarp at iteration %d\n", prev - this, i);
prev = this;
}
pthread_exit(NULL);
}
int main()
{
pthread_t th[NUM_THREADS];
long rc, i;
pid_t pgid;
for (i = 0; i < NUM_THREADS; i++) {
rc = pthread_create(&th[i], NULL, pound, (void *)i);
if (rc < 0)
perror("pthread_create");
}
pthread_exit(NULL);
return 0;
}
==== END pound_clock_gettime.c ====
Suggested-by: Mike Galbraith <mgalbraith@suse.de>
Signed-off-by: Giovanni Gherdovich <ggherdovich@suse.cz>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stanislaw Gruszka <sgruszka@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1470385316-15027-2-git-send-email-ggherdovich@suse.cz
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull scheduler updates from Ingo Molnar:
- introduce and use task_rcu_dereference()/try_get_task_struct() to fix
and generalize task_struct handling (Oleg Nesterov)
- do various per entity load tracking (PELT) fixes and optimizations
(Peter Zijlstra)
- cputime virt-steal time accounting enhancements/fixes (Wanpeng Li)
- introduce consolidated cputime output file cpuacct.usage_all and
related refactorings (Zhao Lei)
- ... plus misc fixes and enhancements
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/core: Panic on scheduling while atomic bugs if kernel.panic_on_warn is set
sched/cpuacct: Introduce cpuacct.usage_all to show all CPU stats together
sched/cpuacct: Use loop to consolidate code in cpuacct_stats_show()
sched/cpuacct: Merge cpuacct_usage_index and cpuacct_stat_index enums
sched/fair: Rework throttle_count sync
sched/core: Fix sched_getaffinity() return value kerneldoc comment
sched/fair: Reorder cgroup creation code
sched/fair: Apply more PELT fixes
sched/fair: Fix PELT integrity for new tasks
sched/cgroup: Fix cpu_cgroup_fork() handling
sched/fair: Fix PELT integrity for new groups
sched/fair: Fix and optimize the fork() path
sched/cputime: Add steal time support to full dynticks CPU time accounting
sched/cputime: Fix prev steal time accouting during CPU hotplug
KVM: Fix steal clock warp during guest CPU hotplug
sched/debug: Always show 'nr_migrations'
sched/fair: Use task_rcu_dereference()
sched/api: Introduce task_rcu_dereference() and try_get_task_struct()
sched/idle: Optimize the generic idle loop
sched/fair: Fix the wrong throttled clock time for cfs_rq_clock_task()
Pull locking updates from Ingo Molnar:
"The locking tree was busier in this cycle than the usual pattern - a
couple of major projects happened to coincide.
The main changes are:
- implement the atomic_fetch_{add,sub,and,or,xor}() API natively
across all SMP architectures (Peter Zijlstra)
- add atomic_fetch_{inc/dec}() as well, using the generic primitives
(Davidlohr Bueso)
- optimize various aspects of rwsems (Jason Low, Davidlohr Bueso,
Waiman Long)
- optimize smp_cond_load_acquire() on arm64 and implement LSE based
atomic{,64}_fetch_{add,sub,and,andnot,or,xor}{,_relaxed,_acquire,_release}()
on arm64 (Will Deacon)
- introduce smp_acquire__after_ctrl_dep() and fix various barrier
mis-uses and bugs (Peter Zijlstra)
- after discovering ancient spin_unlock_wait() barrier bugs in its
implementation and usage, strengthen its semantics and update/fix
usage sites (Peter Zijlstra)
- optimize mutex_trylock() fastpath (Peter Zijlstra)
- ... misc fixes and cleanups"
* 'locking-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (67 commits)
locking/atomic: Introduce inc/dec variants for the atomic_fetch_$op() API
locking/barriers, arch/arm64: Implement LDXR+WFE based smp_cond_load_acquire()
locking/static_keys: Fix non static symbol Sparse warning
locking/qspinlock: Use __this_cpu_dec() instead of full-blown this_cpu_dec()
locking/atomic, arch/tile: Fix tilepro build
locking/atomic, arch/m68k: Remove comment
locking/atomic, arch/arc: Fix build
locking/Documentation: Clarify limited control-dependency scope
locking/atomic, arch/rwsem: Employ atomic_long_fetch_add()
locking/atomic, arch/qrwlock: Employ atomic_fetch_add_acquire()
locking/atomic, arch/mips: Convert to _relaxed atomics
locking/atomic, arch/alpha: Convert to _relaxed atomics
locking/atomic: Remove the deprecated atomic_{set,clear}_mask() functions
locking/atomic: Remove linux/atomic.h:atomic_fetch_or()
locking/atomic: Implement atomic{,64,_long}_fetch_{add,sub,and,andnot,or,xor}{,_relaxed,_acquire,_release}()
locking/atomic: Fix atomic64_relaxed() bits
locking/atomic, arch/xtensa: Implement atomic_fetch_{add,sub,and,or,xor}()
locking/atomic, arch/x86: Implement atomic{,64}_fetch_{add,sub,and,or,xor}()
locking/atomic, arch/tile: Implement atomic{,64}_fetch_{add,sub,and,or,xor}()
locking/atomic, arch/sparc: Implement atomic{,64}_fetch_{add,sub,and,or,xor}()
...
The move of calc_load_migrate() from CPU_DEAD to CPU_DYING did not take into
account that the function is now called from a thread running on the outgoing
CPU. As a result a cpu unplug leakes a load of 1 into the global load
accounting mechanism.
Fix it by adjusting for the currently running thread which calls
calc_load_migrate().
Reported-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Cc: rt@linutronix.de
Cc: shreyas@linux.vnet.ibm.com
Fixes: e9cd8fa4fc: ("sched/migration: Move calc_load_migrate() into CPU_DYING")
Link: http://lkml.kernel.org/r/alpine.DEB.2.11.1607121744350.4083@nanos
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Currently, a schedule while atomic error prints the stack trace to the
kernel log and the system continue running.
Although it is possible to collect the kernel log messages and analyze
it, often more information are needed. Furthermore, keep the system
running is not always the best choice. For example, when the preempt
count underflows the system will not stop to complain about scheduling
while atomic, so the kernel log can wrap around overwriting the first
stack trace, tuning the analysis even more challenging.
This patch uses the kernel.panic_on_warn sysctl to help out on these
more complex situations.
When kernel.panic_on_warn is set to 1, the kernel will panic() in the
schedule while atomic detection.
The default value of the sysctl is 0, maintaining the current behavior.
Signed-off-by: Daniel Bristot de Oliveira <bristot@redhat.com>
Reviewed-by: Luis Claudio R. Goncalves <lgoncalv@redhat.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Luis Claudio R. Goncalves <lgoncalv@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/e8f7b80f353aa22c63bd8557208163989af8493d.1464983675.git.bristot@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Previous version was probably written referencing the man page for
glibc's wrapper, but the wrapper's behavior differs from that of the
syscall itself in this case.
Signed-off-by: Zev Weiss <zev@bewilderbeest.net>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Link: http://lkml.kernel.org/r/1466975603-25408-1-git-send-email-zev@bewilderbeest.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
A future patch needs rq->lock held _after_ we link the task_group into
the hierarchy. In order to avoid taking every rq->lock twice, reorder
things a little and create online_fair_sched_group() to be called
after we link the task_group.
All this code is still ran from css_alloc() so css_online() isn't in
fact used for this.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: bsegall@google.com
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Vincent and Yuyang found another few scenarios in which entity
tracking goes wobbly.
The scenarios are basically due to the fact that new tasks are not
immediately attached and thereby differ from the normal situation -- a
task is always attached to a cfs_rq load average (such that it
includes its blocked contribution) and are explicitly
detached/attached on migration to another cfs_rq.
Scenario 1: switch to fair class
p->sched_class = fair_class;
if (queued)
enqueue_task(p);
...
enqueue_entity()
enqueue_entity_load_avg()
migrated = !sa->last_update_time (true)
if (migrated)
attach_entity_load_avg()
check_class_changed()
switched_from() (!fair)
switched_to() (fair)
switched_to_fair()
attach_entity_load_avg()
If @p is a new task that hasn't been fair before, it will have
!last_update_time and, per the above, end up in
attach_entity_load_avg() _twice_.
Scenario 2: change between cgroups
sched_move_group(p)
if (queued)
dequeue_task()
task_move_group_fair()
detach_task_cfs_rq()
detach_entity_load_avg()
set_task_rq()
attach_task_cfs_rq()
attach_entity_load_avg()
if (queued)
enqueue_task();
...
enqueue_entity()
enqueue_entity_load_avg()
migrated = !sa->last_update_time (true)
if (migrated)
attach_entity_load_avg()
Similar as with scenario 1, if @p is a new task, it will have
!load_update_time and we'll end up in attach_entity_load_avg()
_twice_.
Furthermore, notice how we do a detach_entity_load_avg() on something
that wasn't attached to begin with.
As stated above; the problem is that the new task isn't yet attached
to the load tracking and thereby violates the invariant assumption.
This patch remedies this by ensuring a new task is indeed properly
attached to the load tracking on creation, through
post_init_entity_util_avg().
Of course, this isn't entirely as straightforward as one might think,
since the task is hashed before we call wake_up_new_task() and thus
can be poked at. We avoid this by adding TASK_NEW and teaching
cpu_cgroup_can_attach() to refuse such tasks.
Reported-by: Yuyang Du <yuyang.du@intel.com>
Reported-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
A new fair task is detached and attached from/to task_group with:
cgroup_post_fork()
ss->fork(child) := cpu_cgroup_fork()
sched_move_task()
task_move_group_fair()
Which is wrong, because at this point in fork() the task isn't fully
initialized and it cannot 'move' to another group, because its not
attached to any group as yet.
In fact, cpu_cgroup_fork() needs a small part of sched_move_task() so we
can just call this small part directly instead sched_move_task(). And
the task doesn't really migrate because it is not yet attached so we
need the following sequence:
do_fork()
sched_fork()
__set_task_cpu()
cgroup_post_fork()
set_task_rq() # set task group and runqueue
wake_up_new_task()
select_task_rq() can select a new cpu
__set_task_cpu
post_init_entity_util_avg
attach_task_cfs_rq()
activate_task
enqueue_task
This patch makes that happen.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
[ Added TASK_SET_GROUP to set depth properly. ]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The task_fork_fair() callback already calls __set_task_cpu() and takes
rq->lock.
If we move the sched_class::task_fork callback in sched_fork() under
the existing p->pi_lock, right after its set_task_cpu() call, we can
avoid doing two such calls and omit the IRQ disabling on the rq->lock.
Change to __set_task_cpu() to skip the migration bits, this is a new
task, not a migration. Similarly, make wake_up_new_task() use
__set_task_cpu() for the same reason, the task hasn't actually
migrated as it hasn't ever ran.
This cures the problem of calling migrate_task_rq_fair(), which does
remove_entity_from_load_avg() on tasks that have never been added to
the load avg to begin with.
This bug would result in transiently messed up load_avg values, averaged
out after a few dozen milliseconds. This is probably the reason why
this bug was not found for such a long time.
Reported-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
During CPU hotplug, CPU_ONLINE callbacks are run while the CPU is
online but not active. A CPU_ONLINE callback may create or bind a
kthread so that its cpus_allowed mask only allows the CPU which is
being brought online. The kthread may start executing before the CPU
is made active and can end up in select_fallback_rq().
In such cases, the expected behavior is selecting the CPU which is
coming online; however, because select_fallback_rq() only chooses from
active CPUs, it determines that the task doesn't have any viable CPU
in its allowed mask and ends up overriding it to cpu_possible_mask.
CPU_ONLINE callbacks should be able to put kthreads on the CPU which
is coming online. Update select_fallback_rq() so that it follows
cpu_online() rather than cpu_active() for kthreads.
Reported-by: Gautham R Shenoy <ego@linux.vnet.ibm.com>
Tested-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Abdul Haleem <abdhalee@linux.vnet.ibm.com>
Cc: Aneesh Kumar <aneesh.kumar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: kernel-team@fb.com
Cc: linuxppc-dev@lists.ozlabs.org
Link: http://lkml.kernel.org/r/20160616193504.GB3262@mtj.duckdns.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Lengthy output of sysrq-w may take a lot of time on slow serial console.
Currently we reset NMI-watchdog on the current CPU to avoid spurious
lockup messages. Sometimes this doesn't work since softlockup watchdog
might trigger on another CPU which is waiting for an IPI to proceed.
We reset softlockup watchdogs on all CPUs, but we do this only after
listing all tasks, and this may be too late on a busy system.
So, reset watchdogs CPUs earlier, in for_each_process_thread() loop.
Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/1465474805-14641-1-git-send-email-aryabinin@virtuozzo.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This new form allows using hardware assisted waiting.
Some hardware (ARM64 and x86) allow monitoring an address for changes,
so by providing a pointer we can use this to replace the cpu_relax()
with hardware optimized methods in the future.
Requested-by: Will Deacon <will.deacon@arm.com>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Commit:
e9532e69b8 ("sched/cputime: Fix steal time accounting vs. CPU hotplug")
... set rq->prev_* to 0 after a CPU hotplug comes back, in order to
fix the case where (after CPU hotplug) steal time is smaller than
rq->prev_steal_time.
However, this should never happen. Steal time was only smaller because of the
KVM-specific bug fixed by the previous patch. Worse, the previous patch
triggers a bug on CPU hot-unplug/plug operation: because
rq->prev_steal_time is cleared, all of the CPU's past steal time will be
accounted again on hot-plug.
Since the root cause has been fixed, we can just revert commit e9532e69b8.
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 'commit e9532e69b8 ("sched/cputime: Fix steal time accounting vs. CPU hotplug")'
Link: http://lkml.kernel.org/r/1465813966-3116-3-git-send-email-wanpeng.li@hotmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Merge filesystem stacking fixes from Jann Horn.
* emailed patches from Jann Horn <jannh@google.com>:
sched: panic on corrupted stack end
ecryptfs: forbid opening files without mmap handler
proc: prevent stacking filesystems on top
Until now, hitting this BUG_ON caused a recursive oops (because oops
handling involves do_exit(), which calls into the scheduler, which in
turn raises an oops), which caused stuff below the stack to be
overwritten until a panic happened (e.g. via an oops in interrupt
context, caused by the overwritten CPU index in the thread_info).
Just panic directly.
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The 'schedstats=enable' option doesn't work, and also produces the
following warning during boot:
WARNING: CPU: 0 PID: 0 at /home/jpoimboe/git/linux/kernel/jump_label.c:61 static_key_slow_inc+0x8c/0xa0
static_key_slow_inc used before call to jump_label_init
Modules linked in:
CPU: 0 PID: 0 Comm: swapper Not tainted 4.7.0-rc1+ #25
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.8.1-20150318_183358- 04/01/2014
0000000000000086 3ae3475a4bea95d4 ffffffff81e03da8 ffffffff8143fc83
ffffffff81e03df8 0000000000000000 ffffffff81e03de8 ffffffff810b1ffb
0000003d00000096 ffffffff823514d0 ffff88007ff197c8 0000000000000000
Call Trace:
[<ffffffff8143fc83>] dump_stack+0x85/0xc2
[<ffffffff810b1ffb>] __warn+0xcb/0xf0
[<ffffffff810b207f>] warn_slowpath_fmt+0x5f/0x80
[<ffffffff811e9c0c>] static_key_slow_inc+0x8c/0xa0
[<ffffffff810e07c6>] static_key_enable+0x16/0x40
[<ffffffff8216d633>] setup_schedstats+0x29/0x94
[<ffffffff82148a05>] unknown_bootoption+0x89/0x191
[<ffffffff810d8617>] parse_args+0x297/0x4b0
[<ffffffff82148d61>] start_kernel+0x1d8/0x4a9
[<ffffffff8214897c>] ? set_init_arg+0x55/0x55
[<ffffffff82148120>] ? early_idt_handler_array+0x120/0x120
[<ffffffff821482db>] x86_64_start_reservations+0x2f/0x31
[<ffffffff82148427>] x86_64_start_kernel+0x14a/0x16d
The problem is that it tries to update the 'sched_schedstats' static key
before jump labels have been initialized.
Changing jump_label_init() to be called earlier before
parse_early_param() wouldn't fix it: it would still fail trying to
poke_text() because mm isn't yet initialized.
Instead, just create a temporary '__sched_schedstats' variable which can
be copied to the static key later during sched_init() after jump labels
have been initialized.
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matt Fleming <matt@codeblueprint.co.uk>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: cb2517653f ("sched/debug: Make schedstats a runtime tunable that is disabled by default")
Link: http://lkml.kernel.org/r/453775fe3433bed65731a583e228ccea806d18cd.1465322027.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Commit:
b5179ac70d ("sched/fair: Prepare to fix fairness problems on migration")
... introduced a bug: Mike Galbraith found that it introduced a
performance regression, while Paul E. McKenney reported lost
wakeups and bisected it to this commit.
The reason is that I mis-read ttwu_queue() such that I assumed any
wakeup that got a remote queue must have had the task migrated.
Since this is not so; we need to transfer this information between
queueing the wakeup and actually doing the wakeup. Use a new
task_struct::sched_flag for this, we already write to
sched_contributes_to_load in the wakeup path so this is a hot and
modified cacheline.
Reported-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reported-by: Mike Galbraith <umgwanakikbuti@gmail.com>
Tested-by: Mike Galbraith <umgwanakikbuti@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Hunter <ahh@google.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Ben Segall <bsegall@google.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matt Fleming <matt@codeblueprint.co.uk>
Cc: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Paul Turner <pjt@google.com>
Cc: Pavan Kondeti <pkondeti@codeaurora.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Quentin Casasnovas <quentin.casasnovas@oracle.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: byungchul.park@lge.com
Fixes: b5179ac70d ("sched/fair: Prepare to fix fairness problems on migration")
Link: http://lkml.kernel.org/r/20160523091907.GD15728@worktop.ger.corp.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
tsk_nr_cpus_allowed() is an accessor for task->nr_cpus_allowed which allows
us to change the representation of ->nr_cpus_allowed if required.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: linux-kernel@vger.kernel.org
Link: http://lkml.kernel.org/r/1462969411-17735-2-git-send-email-bigeasy@linutronix.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The following commit:
9642d18eee ("nohz: Affine unpinned timers to housekeepers")'
intended to affine unpinned timers to housekeepers:
unpinned timers(full dynaticks, idle) => nearest busy housekeepers(otherwise, fallback to any housekeepers)
unpinned timers(full dynaticks, busy) => nearest busy housekeepers(otherwise, fallback to any housekeepers)
unpinned timers(houserkeepers, idle) => nearest busy housekeepers(otherwise, fallback to itself)
However, the !idle_cpu(i) && is_housekeeping_cpu(cpu) check modified the
intention to:
unpinned timers(full dynaticks, idle) => any housekeepers(no mattter cpu topology)
unpinned timers(full dynaticks, busy) => any housekeepers(no mattter cpu topology)
unpinned timers(housekeepers, idle) => any busy cpus(otherwise, fallback to any housekeepers)
This patch fixes it by checking if there are busy housekeepers nearby,
otherwise falls to any housekeepers/itself. After the patch:
unpinned timers(full dynaticks, idle) => nearest busy housekeepers(otherwise, fallback to any housekeepers)
unpinned timers(full dynaticks, busy) => nearest busy housekeepers(otherwise, fallback to any housekeepers)
unpinned timers(housekeepers, idle) => nearest busy housekeepers(otherwise, fallback to itself)
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
[ Fixed the changelog. ]
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Fixes: 'commit 9642d18eee ("nohz: Affine unpinned timers to housekeepers")'
Link: http://lkml.kernel.org/r/1462344334-8303-1-git-send-email-wanpeng.li@hotmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
With sched_class::task_waking being called only when we do
set_task_cpu(), we can make sched_class::migrate_task_rq() do the work
and eliminate sched_class::task_waking entirely.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Hunter <ahh@google.com>
Cc: Ben Segall <bsegall@google.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matt Fleming <matt@codeblueprint.co.uk>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Mike Galbraith <umgwanakikbuti@gmail.com>
Cc: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: Paul Turner <pjt@google.com>
Cc: Pavan Kondeti <pkondeti@codeaurora.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: byungchul.park@lge.com
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Mike reported that our recent attempt to fix migration problems:
3a47d5124a ("sched/fair: Fix fairness issue on migration")
broke interactivity and the signal starve test. We reverted that
commit and now let's try it again more carefully, with some other
underlying problems fixed first.
One problem is that I assumed ENQUEUE_WAKING was only set when we do a
cross-cpu wakeup (migration), which isn't true. This means we now
destroy the vruntime history of tasks and wakeup-preemption suffers.
Cure this by making my assumption true, only call
sched_class::task_waking() when we do a cross-cpu wakeup. This avoids
the indirect call in the case we do a local wakeup.
Reported-by: Mike Galbraith <mgalbraith@suse.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Hunter <ahh@google.com>
Cc: Ben Segall <bsegall@google.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matt Fleming <matt@codeblueprint.co.uk>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: Paul Turner <pjt@google.com>
Cc: Pavan Kondeti <pkondeti@codeaurora.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: byungchul.park@lge.com
Cc: linux-kernel@vger.kernel.org
Fixes: 3a47d5124a ("sched/fair: Fix fairness issue on migration")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
No need for an extra notifier. We don't need to handle all these states. It's
sufficient to kill the timer when the cpu dies.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: rt@linutronix.de
Link: http://lkml.kernel.org/r/20160310120025.770528462@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The alleged requirement that the migration notifier has a lower priority than
perf is completely undocumented and there is no indication at all that this is
true. perf does not even handle the CPU_ONLINE notification and perf really
has nothing to do with migration.
Move the CPU_ONLINE code into the sched_activate_cpu() state callback.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: rt@linutronix.de
Link: http://lkml.kernel.org/r/20160310120025.421743581@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
It really does not matter when we fold the load for the outgoing cpu. It's
almost dead anyway, so there is no harm if we fail to fold the few
microseconds which are required for going fully away.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: rt@linutronix.de
Link: http://lkml.kernel.org/r/20160310120025.328739226@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
We can piggy pack that on the SCHED_STARTING state. It's not required before
the cpu actually comes online. Name the function proper as it has nothing to
do with migration.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: rt@linutronix.de
Link: http://lkml.kernel.org/r/20160310120025.248226511@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The sync_rcu stuff is specificically for clearing bits in the active
mask, such that everybody will observe the bit cleared and will not
consider the cleared CPU for load-balancing etc.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: rt@linutronix.de
Link: http://lkml.kernel.org/r/20160310120025.169219710@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Now that we reduced everything into single notifiers, it's simple to move them
into the hotplug state machine space.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: rt@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This is the last operation on the cpu before vanishing. No point in calling
that on CPU_DEAD.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: rt@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
We can maintain the ordering of the scheduler cpu hotplug functionality nicely
in one notifer. Get rid of the maze.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: rt@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Prevent the SMP scheduler related notifiers to be executed before the smp
scheduler is initialized and install them early.
This is a preparatory change for further consolidation of the hotplug notifier
maze.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: rt@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Start distangling the maze of hotplug notifiers in the scheduler.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: rt@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
In order to enable symmetric hotplug, we must mirror the online &&
!active state of cpu-down on the cpu-up side.
However, to retain sanity, limit this state to per-cpu kthreads.
Aside from the change to set_cpus_allowed_ptr(), which allow moving
the per-cpu kthreads on, the other critical piece is the cpu selection
for pinned tasks in select_task_rq(). This avoids dropping into
select_fallback_rq().
select_fallback_rq() cannot be allowed to select !active cpus because
its used to migrate user tasks away. And we do not want to move user
tasks onto cpus that are in transition.
Requested-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Jan H. Schönherr <jschoenh@amazon.de>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: rt@linutronix.de
Link: http://lkml.kernel.org/r/20160301152303.GV6356@twins.programming.kicks-ass.net
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The problem with the existing lock pinning is that each pin is of
value 1; this mean you can simply unpin if you know its pinned,
without having any extra information.
This scheme generates a random (16 bit) cookie for each pin and
requires this same cookie to unpin. This means you have to keep the
cookie in context.
No objsize difference for !LOCKDEP kernels.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In order to be able to pass around more than just the IRQ flags in the
future, add a rq_flags structure.
No difference in code generation for the x86_64-defconfig build I
tested.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
By default, this is the same thing as switch_mm().
x86 will override it as an optimization.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Reviewed-by: Borislav Petkov <bp@suse.de>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/df401df47bdd6be3e389c6f1e3f5310d70e81b2c.1461688545.git.luto@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Chris Metcalf reported a that sched_can_stop_tick() sometimes fails to
re-enable the tick.
His observed problem is that rq->cfs.nr_running can be 1 even though
there are multiple runnable CFS tasks. This happens in the cgroup
case, in which case cfs.nr_running is the number of runnable entities
for that level.
If there is a single runnable cgroup (which can have an arbitrary
number of runnable child entries itself) rq->cfs.nr_running will be 1.
However, looking at that function I think there's more problems with it.
It seems to assume that if there's FIFO tasks, those will run. This is
incorrect. The FIFO task can have a lower prio than an RR task, in which
case the RR task will run.
So the whole fifo_nr_running test seems misplaced, it should go after
the rr_nr_running tests. That is, only if !rr_nr_running, can we use
fifo_nr_running like this.
Reported-by: Chris Metcalf <cmetcalf@mellanox.com>
Tested-by: Chris Metcalf <cmetcalf@mellanox.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Luiz Capitulino <lcapitulino@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Cc: Wanpeng Li <kernellwp@gmail.com>
Fixes: 76d92ac305 ("sched: Migrate sched to use new tick dependency mask model")
Link: http://lkml.kernel.org/r/20160421160315.GK24771@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
I got a minus(very big) dl_b->total_bw during my deadline tests.
# grep dl /proc/sched_debug
dl_rq[0]:
.dl_nr_running : 0
.dl_bw->bw : 996147
.dl_bw->total_bw : -222297900
Something unusual must have happened.
After some digging, I finally noticed that when changing a deadline
task to normal(cfs), and changing it back to deadline immediately,
after it died, we will got the wrong dl_bw->total_bw.
The root cause is in dl_overflow(), it has:
if (new_bw == p->dl.dl_bw)
return 0;
1) When a deadline task is changed to !deadline task, it will start
dl timer in switched_from_dl(), and retain previous deadline parameter
till the timer expires.
2) If we change it back to deadline with the same bandwidth parameter
before the timer expires, as it keeps the old bandwidth although it
is not a deadline task. dl_overflow() simply returns success without
updating the right data, and got the wrong dl_bw->total_bw.
The solution is simple, if @p is not deadline, don't return.
Signed-off-by: Xunlei Pang <xlpang@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Juri Lelli <juri.lelli@arm.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1460636368-1993-1-git-send-email-xlpang@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Some code in CPU load update only concern NO_HZ configs but it is
built on all configurations. When NO_HZ isn't built, that code is harmless
but just happens to take some useless ressources in CPU and memory:
1) one useless field in struct rq
2) jiffies record on every tick that is never used (cpu_load_update_periodic)
3) decay_load_missed is called two times on every tick to eventually
return immediately with no action taken. And that function is dead
code.
For pure optimization purposes, lets conditionally build the NO_HZ
related code.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Byungchul Park <byungchul.park@lge.com>
Cc: Chris Metcalf <cmetcalf@ezchip.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Luiz Capitulino <lcapitulino@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul E . McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1461080211-16271-1-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The CPU load update related functions have a weak naming convention
currently, starting with update_cpu_load_*() which isn't ideal as
"update" is a very generic concept.
Since two of these functions are public already (and a third is to come)
that's enough to introduce a more conventional naming scheme. So let's
do the following rename instead:
update_cpu_load_*() -> cpu_load_update_*()
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Byungchul Park <byungchul.park@lge.com>
Cc: Chris Metcalf <cmetcalf@ezchip.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Luiz Capitulino <lcapitulino@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul E . McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1460555812-25375-2-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
sysrq_sched_debug_show() can dump a lot of information. Don't print out
all that if we're just trying to get a list of blocked tasks (SysRq-W).
The information is still accessible with SysRq-T.
Signed-off-by: Rabin Vincent <rabinv@axis.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1459777322-30902-1-git-send-email-rabin.vincent@axis.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
A new task's util_avg is set to full utilization of a CPU (100% time
running). This accelerates a new task's utilization ramp-up, useful to
boost its execution in early time. However, it may result in
(insanely) high utilization for a transient time period when a flood
of tasks are spawned. Importantly, it violates the "fundamentally
bounded" CPU utilization, and its side effect is negative if we don't
take any measure to bound it.
This patch proposes an algorithm to address this issue. It has
two methods to approach a sensible initial util_avg:
(1) An expected (or average) util_avg based on its cfs_rq's util_avg:
util_avg = cfs_rq->util_avg / (cfs_rq->load_avg + 1) * se.load.weight
(2) A trajectory of how successive new tasks' util develops, which
gives 1/2 of the left utilization budget to a new task such that
the additional util is noticeably large (when overall util is low) or
unnoticeably small (when overall util is high enough). In the meantime,
the aggregate utilization is well bounded:
util_avg_cap = (1024 - cfs_rq->avg.util_avg) / 2^n
where n denotes the nth task.
If util_avg is larger than util_avg_cap, then the effective util is
clamped to the util_avg_cap.
Reported-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Signed-off-by: Yuyang Du <yuyang.du@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: bsegall@google.com
Cc: morten.rasmussen@arm.com
Cc: pjt@google.com
Cc: steve.muckle@linaro.org
Link: http://lkml.kernel.org/r/1459283456-21682-1-git-send-email-yuyang.du@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
While testing the tracer preemptoff, I hit this strange trace:
<...>-259 0...1 0us : schedule <-worker_thread
<...>-259 0d..1 0us : rcu_note_context_switch <-__schedule
<...>-259 0d..1 0us : rcu_sched_qs <-rcu_note_context_switch
<...>-259 0d..1 0us : rcu_preempt_qs <-rcu_note_context_switch
<...>-259 0d..1 0us : _raw_spin_lock <-__schedule
<...>-259 0d..1 0us : preempt_count_add <-_raw_spin_lock
<...>-259 0d..2 0us : do_raw_spin_lock <-_raw_spin_lock
<...>-259 0d..2 1us : deactivate_task <-__schedule
<...>-259 0d..2 1us : update_rq_clock.part.84 <-deactivate_task
<...>-259 0d..2 1us : dequeue_task_fair <-deactivate_task
<...>-259 0d..2 1us : dequeue_entity <-dequeue_task_fair
<...>-259 0d..2 1us : update_curr <-dequeue_entity
<...>-259 0d..2 1us : update_min_vruntime <-update_curr
<...>-259 0d..2 1us : cpuacct_charge <-update_curr
<...>-259 0d..2 1us : __rcu_read_lock <-cpuacct_charge
<...>-259 0d..2 1us : __rcu_read_unlock <-cpuacct_charge
<...>-259 0d..2 1us : clear_buddies <-dequeue_entity
<...>-259 0d..2 1us : account_entity_dequeue <-dequeue_entity
<...>-259 0d..2 2us : update_min_vruntime <-dequeue_entity
<...>-259 0d..2 2us : update_cfs_shares <-dequeue_entity
<...>-259 0d..2 2us : hrtick_update <-dequeue_task_fair
<...>-259 0d..2 2us : wq_worker_sleeping <-__schedule
<...>-259 0d..2 2us : kthread_data <-wq_worker_sleeping
<...>-259 0d..2 2us : pick_next_task_fair <-__schedule
<...>-259 0d..2 2us : check_cfs_rq_runtime <-pick_next_task_fair
<...>-259 0d..2 2us : pick_next_entity <-pick_next_task_fair
<...>-259 0d..2 2us : clear_buddies <-pick_next_entity
<...>-259 0d..2 2us : pick_next_entity <-pick_next_task_fair
<...>-259 0d..2 2us : clear_buddies <-pick_next_entity
<...>-259 0d..2 2us : set_next_entity <-pick_next_task_fair
<...>-259 0d..2 3us : put_prev_entity <-pick_next_task_fair
<...>-259 0d..2 3us : check_cfs_rq_runtime <-put_prev_entity
<...>-259 0d..2 3us : set_next_entity <-pick_next_task_fair
gnome-sh-1031 0d..2 3us : finish_task_switch <-__schedule
gnome-sh-1031 0d..2 3us : _raw_spin_unlock_irq <-finish_task_switch
gnome-sh-1031 0d..2 3us : do_raw_spin_unlock <-_raw_spin_unlock_irq
gnome-sh-1031 0...2 3us!: preempt_count_sub <-_raw_spin_unlock_irq
gnome-sh-1031 0...1 582us : do_raw_spin_lock <-_raw_spin_lock
gnome-sh-1031 0...1 583us : _raw_spin_unlock <-drm_gem_object_lookup
gnome-sh-1031 0...1 583us : do_raw_spin_unlock <-_raw_spin_unlock
gnome-sh-1031 0...1 583us : preempt_count_sub <-_raw_spin_unlock
gnome-sh-1031 0...1 584us : _raw_spin_unlock <-drm_gem_object_lookup
gnome-sh-1031 0...1 584us+: trace_preempt_on <-drm_gem_object_lookup
gnome-sh-1031 0...1 603us : <stack trace>
=> preempt_count_sub
=> _raw_spin_unlock
=> drm_gem_object_lookup
=> i915_gem_madvise_ioctl
=> drm_ioctl
=> do_vfs_ioctl
=> SyS_ioctl
=> entry_SYSCALL_64_fastpath
As I'm tracing preemption disabled, it seemed incorrect that the trace
would go across a schedule and report not being in the scheduler.
Looking into this I discovered the problem.
schedule() calls preempt_disable() but the preempt_schedule() calls
preempt_enable_notrace(). What happened above was that the gnome-shell
task was preempted on another CPU, migrated over to the idle cpu. The
tracer stared with idle calling schedule(), which called
preempt_disable(), but then gnome-shell finished, and it enabled
preemption with preempt_enable_notrace() that does stop the trace, even
though preemption was enabled.
The purpose of the preempt_disable_notrace() in the preempt_schedule()
is to prevent function tracing from going into an infinite loop.
Because function tracing can trace the preempt_enable/disable() calls
that are traced. The problem with function tracing is:
NEED_RESCHED set
preempt_schedule()
preempt_disable()
preempt_count_inc()
function trace (before incrementing preempt count)
preempt_disable_notrace()
preempt_enable_notrace()
sees NEED_RESCHED set
preempt_schedule() (repeat)
Now by breaking out the preempt off/on tracing into their own code:
preempt_disable_check() and preempt_enable_check(), we can add these to
the preempt_schedule() code. As preemption would then be disabled, even
if they were to be traced by the function tracer, the disabled
preemption would prevent the recursion.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20160321112339.6dc78ad6@gandalf.local.home
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This patch functionally reverts:
5fd7a09cfb ("atomic: Export fetch_or()")
During the merge Linus observed that the generic version of fetch_or()
was messy:
" This makes the ugly "fetch_or()" macro that the scheduler used
internally a new generic helper, and does a bad job at it. "
e23604edac Merge branch 'timers-nohz-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Now that we have introduced atomic_fetch_or(), fetch_or() is only used
by the scheduler in order to deal with thread_info flags which type
can vary across architectures.
Lets confine fetch_or() back to the scheduler so that we encourage
future users to use the more robust and well typed atomic_t version
instead.
While at it, fetch_or() gets robustified, pasting improvements from a
previous patch by Ingo Molnar that avoids needless expression
re-evaluations in the loop.
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1458830281-4255-4-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull scheduler fixes from Ingo Molnar:
"Misc fixes: a cgroup fix, a fair-scheduler migration accounting fix, a
cputime fix and two cpuacct cleanups"
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/cpuacct: Simplify the cpuacct code
sched/cpuacct: Rename parameter in cpuusage_write() for readability
sched/fair: Add comments to explain select_idle_sibling()
sched/fair: Fix fairness issue on migration
sched/cgroup: Fix/cleanup cgroup teardown/init
sched/cputime: Fix steal time accounting vs. CPU hotplug
The CPU controller hasn't kept up with the various changes in the whole
cgroup initialization / destruction sequence, and commit:
2e91fa7f6d ("cgroup: keep zombies associated with their original cgroups")
caused it to explode.
The reason for this is that zombies do not inhibit css_offline() from
being called, but do stall css_released(). Now we tear down the cfs_rq
structures on css_offline() but zombies can run after that, leading to
use-after-free issues.
The solution is to move the tear-down to css_released(), which
guarantees nobody (including no zombies) is still using our cgroup.
Furthermore, a few simple cleanups are possible too. There doesn't
appear to be any point to us using css_online() (anymore?) so fold that
in css_alloc().
And since cgroup code guarantees an RCU grace period between
css_released() and css_free() we can forgo using call_rcu() and free the
stuff immediately.
Suggested-by: Tejun Heo <tj@kernel.org>
Reported-by: Kazuki Yamaguchi <k@rhe.jp>
Reported-by: Niklas Cassel <niklas.cassel@axis.com>
Tested-by: Niklas Cassel <niklas.cassel@axis.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 2e91fa7f6d ("cgroup: keep zombies associated with their original cgroups")
Link: http://lkml.kernel.org/r/20160316152245.GY6344@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull 'objtool' stack frame validation from Ingo Molnar:
"This tree adds a new kernel build-time object file validation feature
(ONFIG_STACK_VALIDATION=y): kernel stack frame correctness validation.
It was written by and is maintained by Josh Poimboeuf.
The motivation: there's a category of hard to find kernel bugs, most
of them in assembly code (but also occasionally in C code), that
degrades the quality of kernel stack dumps/backtraces. These bugs are
hard to detect at the source code level. Such bugs result in
incorrect/incomplete backtraces most of time - but can also in some
rare cases result in crashes or other undefined behavior.
The build time correctness checking is done via the new 'objtool'
user-space utility that was written for this purpose and which is
hosted in the kernel repository in tools/objtool/. The tool's (very
simple) UI and source code design is shaped after Git and perf and
shares quite a bit of infrastructure with tools/perf (which tooling
infrastructure sharing effort got merged via perf and is already
upstream). Objtool follows the well-known kernel coding style.
Objtool does not try to check .c or .S files, it instead analyzes the
resulting .o generated machine code from first principles: it decodes
the instruction stream and interprets it. (Right now objtool supports
the x86-64 architecture.)
From tools/objtool/Documentation/stack-validation.txt:
"The kernel CONFIG_STACK_VALIDATION option enables a host tool named
objtool which runs at compile time. It has a "check" subcommand
which analyzes every .o file and ensures the validity of its stack
metadata. It enforces a set of rules on asm code and C inline
assembly code so that stack traces can be reliable.
Currently it only checks frame pointer usage, but there are plans to
add CFI validation for C files and CFI generation for asm files.
For each function, it recursively follows all possible code paths
and validates the correct frame pointer state at each instruction.
It also follows code paths involving special sections, like
.altinstructions, __jump_table, and __ex_table, which can add
alternative execution paths to a given instruction (or set of
instructions). Similarly, it knows how to follow switch statements,
for which gcc sometimes uses jump tables."
When this new kernel option is enabled (it's disabled by default), the
tool, if it finds any suspicious assembly code pattern, outputs
warnings in compiler warning format:
warning: objtool: rtlwifi_rate_mapping()+0x2e7: frame pointer state mismatch
warning: objtool: cik_tiling_mode_table_init()+0x6ce: call without frame pointer save/setup
warning: objtool:__schedule()+0x3c0: duplicate frame pointer save
warning: objtool:__schedule()+0x3fd: sibling call from callable instruction with changed frame pointer
... so that scripts that pick up compiler warnings will notice them.
All known warnings triggered by the tool are fixed by the tree, most
of the commits in fact prepare the kernel to be warning-free. Most of
them are bugfixes or cleanups that stand on their own, but there are
also some annotations of 'special' stack frames for justified cases
such entries to JIT-ed code (BPF) or really special boot time code.
There are two other long-term motivations behind this tool as well:
- To improve the quality and reliability of kernel stack frames, so
that they can be used for optimized live patching.
- To create independent infrastructure to check the correctness of
CFI stack frames at build time. CFI debuginfo is notoriously
unreliable and we cannot use it in the kernel as-is without extra
checking done both on the kernel side and on the build side.
The quality of kernel stack frames matters to debuggability as well,
so IMO we can merge this without having to consider the live patching
or CFI debuginfo angle"
* 'core-objtool-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (52 commits)
objtool: Only print one warning per function
objtool: Add several performance improvements
tools: Copy hashtable.h into tools directory
objtool: Fix false positive warnings for functions with multiple switch statements
objtool: Rename some variables and functions
objtool: Remove superflous INIT_LIST_HEAD
objtool: Add helper macros for traversing instructions
objtool: Fix false positive warnings related to sibling calls
objtool: Compile with debugging symbols
objtool: Detect infinite recursion
objtool: Prevent infinite recursion in noreturn detection
objtool: Detect and warn if libelf is missing and don't break the build
tools: Support relative directory path for 'O='
objtool: Support CROSS_COMPILE
x86/asm/decoder: Use explicitly signed chars
objtool: Enable stack metadata validation on 64-bit x86
objtool: Add CONFIG_STACK_VALIDATION option
objtool: Add tool to perform compile-time stack metadata validation
x86/kprobes: Mark kretprobe_trampoline() stack frame as non-standard
sched: Always inline context_switch()
...
Pull cgroup updates from Tejun Heo:
"cgroup changes for v4.6-rc1. No userland visible behavior changes in
this pull request. I'll send out a separate pull request for the
addition of cgroup namespace support.
- The biggest change is the revamping of cgroup core task migration
and controller handling logic. There are quite a few places where
controllers and tasks are manipulated. Previously, many of those
places implemented custom operations for each specific use case
assuming specific starting conditions. While this worked, it makes
the code fragile and difficult to follow.
The bulk of this pull request restructures these operations so that
most related operations are performed through common helpers which
implement recursive (subtrees are always processed consistently)
and idempotent (they make cgroup hierarchy converge to the target
state rather than performing operations assuming specific starting
conditions). This makes the code a lot easier to understand,
verify and extend.
- Implicit controller support is added. This is primarily for using
perf_event on the v2 hierarchy so that perf can match cgroup v2
path without requiring the user to do anything special. The kernel
portion of perf_event changes is acked but userland changes are
still pending review.
- cgroup_no_v1= boot parameter added to ease testing cgroup v2 in
certain environments.
- There is a regression introduced during v4.4 devel cycle where
attempts to migrate zombie tasks can mess up internal object
management. This was fixed earlier this week and included in this
pull request w/ stable cc'd.
- Misc non-critical fixes and improvements"
* 'for-4.6' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup: (44 commits)
cgroup: avoid false positive gcc-6 warning
cgroup: ignore css_sets associated with dead cgroups during migration
Documentation: cgroup v2: Trivial heading correction.
cgroup: implement cgroup_subsys->implicit_on_dfl
cgroup: use css_set->mg_dst_cgrp for the migration target cgroup
cgroup: make cgroup[_taskset]_migrate() take cgroup_root instead of cgroup
cgroup: move migration destination verification out of cgroup_migrate_prepare_dst()
cgroup: fix incorrect destination cgroup in cgroup_update_dfl_csses()
cgroup: Trivial correction to reflect controller.
cgroup: remove stale item in cgroup-v1 document INDEX file.
cgroup: update css iteration in cgroup_update_dfl_csses()
cgroup: allocate 2x cgrp_cset_links when setting up a new root
cgroup: make cgroup_calc_subtree_ss_mask() take @this_ss_mask
cgroup: reimplement rebind_subsystems() using cgroup_apply_control() and friends
cgroup: use cgroup_apply_enable_control() in cgroup creation path
cgroup: combine cgroup_mutex locking and offline css draining
cgroup: factor out cgroup_{apply|finalize}_control() from cgroup_subtree_control_write()
cgroup: introduce cgroup_{save|propagate|restore}_control()
cgroup: make cgroup_drain_offline() and cgroup_apply_control_{disable|enable}() recursive
cgroup: factor out cgroup_apply_control_enable() from cgroup_subtree_control_write()
...
Pull workqueue updates from Tejun Heo:
"Three trivial workqueue changes"
* 'for-4.6' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
workqueue: Fix comment for work_on_cpu()
sched/core: Get rid of 'cpu' argument in wq_worker_sleeping()
workqueue: Replace usage of init_name with dev_set_name()
Pull cpu hotplug updates from Thomas Gleixner:
"This is the first part of the ongoing cpu hotplug rework:
- Initial implementation of the state machine
- Runs all online and prepare down callbacks on the plugged cpu and
not on some random processor
- Replaces busy loop waiting with completions
- Adds tracepoints so the states can be followed"
More detailed commentary on this work from an earlier email:
"What's wrong with the current cpu hotplug infrastructure?
- Asymmetry
The hotplug notifier mechanism is asymmetric versus the bringup and
teardown. This is mostly caused by the notifier mechanism.
- Largely undocumented dependencies
While some notifiers use explicitely defined notifier priorities,
we have quite some notifiers which use numerical priorities to
express dependencies without any documentation why.
- Control processor driven
Most of the bringup/teardown of a cpu is driven by a control
processor. While it is understandable, that preperatory steps,
like idle thread creation, memory allocation for and initialization
of essential facilities needs to be done before a cpu can boot,
there is no reason why everything else must run on a control
processor. Before this patch series, bringup looks like this:
Control CPU Booting CPU
do preparatory steps
kick cpu into life
do low level init
sync with booting cpu sync with control cpu
bring the rest up
- All or nothing approach
There is no way to do partial bringups. That's something which is
really desired because we waste e.g. at boot substantial amount of
time just busy waiting that the cpu comes to life. That's stupid
as we could very well do preparatory steps and the initial IPI for
other cpus and then go back and do the necessary low level
synchronization with the freshly booted cpu.
- Minimal debuggability
Due to the notifier based design, it's impossible to switch between
two stages of the bringup/teardown back and forth in order to test
the correctness. So in many hotplug notifiers the cancel
mechanisms are either not existant or completely untested.
- Notifier [un]registering is tedious
To [un]register notifiers we need to protect against hotplug at
every callsite. There is no mechanism that bringup/teardown
callbacks are issued on the online cpus, so every caller needs to
do it itself. That also includes error rollback.
What's the new design?
The base of the new design is a symmetric state machine, where both
the control processor and the booting/dying cpu execute a well
defined set of states. Each state is symmetric in the end, except
for some well defined exceptions, and the bringup/teardown can be
stopped and reversed at almost all states.
So the bringup of a cpu will look like this in the future:
Control CPU Booting CPU
do preparatory steps
kick cpu into life
do low level init
sync with booting cpu sync with control cpu
bring itself up
The synchronization step does not require the control cpu to wait.
That mechanism can be done asynchronously via a worker or some
other mechanism.
The teardown can be made very similar, so that the dying cpu cleans
up and brings itself down. Cleanups which need to be done after
the cpu is gone, can be scheduled asynchronously as well.
There is a long way to this, as we need to refactor the notion when a
cpu is available. Today we set the cpu online right after it comes
out of the low level bringup, which is not really correct.
The proper mechanism is to set it to available, i.e. cpu local
threads, like softirqd, hotplug thread etc. can be scheduled on that
cpu, and once it finished all booting steps, it's set to online, so
general workloads can be scheduled on it. The reverse happens on
teardown. First thing to do is to forbid scheduling of general
workloads, then teardown all the per cpu resources and finally shut it
off completely.
This patch series implements the basic infrastructure for this at the
core level. This includes the following:
- Basic state machine implementation with well defined states, so
ordering and prioritization can be expressed.
- Interfaces to [un]register state callbacks
This invokes the bringup/teardown callback on all online cpus with
the proper protection in place and [un]installs the callbacks in
the state machine array.
For callbacks which have no particular ordering requirement we have
a dynamic state space, so that drivers don't have to register an
explicit hotplug state.
If a callback fails, the code automatically does a rollback to the
previous state.
- Sysfs interface to drive the state machine to a particular step.
This is only partially functional today. Full functionality and
therefor testability will be achieved once we converted all
existing hotplug notifiers over to the new scheme.
- Run all CPU_ONLINE/DOWN_PREPARE notifiers on the booting/dying
processor:
Control CPU Booting CPU
do preparatory steps
kick cpu into life
do low level init
sync with booting cpu sync with control cpu
wait for boot
bring itself up
Signal completion to control cpu
In a previous step of this work we've done a full tree mechanical
conversion of all hotplug notifiers to the new scheme. The balance
is a net removal of about 4000 lines of code.
This is not included in this series, as we decided to take a
different approach. Instead of mechanically converting everything
over, we will do a proper overhaul of the usage sites one by one so
they nicely fit into the symmetric callback scheme.
I decided to do that after I looked at the ugliness of some of the
converted sites and figured out that their hotplug mechanism is
completely buggered anyway. So there is no point to do a
mechanical conversion first as we need to go through the usage
sites one by one again in order to achieve a full symmetric and
testable behaviour"
* 'smp-hotplug-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (23 commits)
cpu/hotplug: Document states better
cpu/hotplug: Fix smpboot thread ordering
cpu/hotplug: Remove redundant state check
cpu/hotplug: Plug death reporting race
rcu: Make CPU_DYING_IDLE an explicit call
cpu/hotplug: Make wait for dead cpu completion based
cpu/hotplug: Let upcoming cpu bring itself fully up
arch/hotplug: Call into idle with a proper state
cpu/hotplug: Move online calls to hotplugged cpu
cpu/hotplug: Create hotplug threads
cpu/hotplug: Split out the state walk into functions
cpu/hotplug: Unpark smpboot threads from the state machine
cpu/hotplug: Move scheduler cpu_online notifier to hotplug core
cpu/hotplug: Implement setup/removal interface
cpu/hotplug: Make target state writeable
cpu/hotplug: Add sysfs state interface
cpu/hotplug: Hand in target state to _cpu_up/down
cpu/hotplug: Convert the hotplugged cpu work to a state machine
cpu/hotplug: Convert to a state machine for the control processor
cpu/hotplug: Add tracepoints
...
Pull NOHZ updates from Ingo Molnar:
"NOHZ enhancements, by Frederic Weisbecker, which reorganizes/refactors
the NOHZ 'can the tick be stopped?' infrastructure and related code to
be data driven, and harmonizes the naming and handling of all the
various properties"
[ This makes the ugly "fetch_or()" macro that the scheduler used
internally a new generic helper, and does a bad job at it.
I'm pulling it, but I've asked Ingo and Frederic to get this
fixed up ]
* 'timers-nohz-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched-clock: Migrate to use new tick dependency mask model
posix-cpu-timers: Migrate to use new tick dependency mask model
sched: Migrate sched to use new tick dependency mask model
sched: Account rr tasks
perf: Migrate perf to use new tick dependency mask model
nohz: Use enum code for tick stop failure tracing message
nohz: New tick dependency mask
nohz: Implement wide kick on top of irq work
atomic: Export fetch_or()
Pull scheduler updates from Ingo Molnar:
"The main changes in this cycle are:
- Make schedstats a runtime tunable (disabled by default) and
optimize it via static keys.
As most distributions enable CONFIG_SCHEDSTATS=y due to its
instrumentation value, this is a nice performance enhancement.
(Mel Gorman)
- Implement 'simple waitqueues' (swait): these are just pure
waitqueues without any of the more complex features of full-blown
waitqueues (callbacks, wake flags, wake keys, etc.). Simple
waitqueues have less memory overhead and are faster.
Use simple waitqueues in the RCU code (in 4 different places) and
for handling KVM vCPU wakeups.
(Peter Zijlstra, Daniel Wagner, Thomas Gleixner, Paul Gortmaker,
Marcelo Tosatti)
- sched/numa enhancements (Rik van Riel)
- NOHZ performance enhancements (Rik van Riel)
- Various sched/deadline enhancements (Steven Rostedt)
- Various fixes (Peter Zijlstra)
- ... and a number of other fixes, cleanups and smaller enhancements"
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (29 commits)
sched/cputime: Fix steal_account_process_tick() to always return jiffies
sched/deadline: Remove dl_new from struct sched_dl_entity
Revert "kbuild: Add option to turn incompatible pointer check into error"
sched/deadline: Remove superfluous call to switched_to_dl()
sched/debug: Fix preempt_disable_ip recording for preempt_disable()
sched, time: Switch VIRT_CPU_ACCOUNTING_GEN to jiffy granularity
time, acct: Drop irq save & restore from __acct_update_integrals()
acct, time: Change indentation in __acct_update_integrals()
sched, time: Remove non-power-of-two divides from __acct_update_integrals()
sched/rt: Kick RT bandwidth timer immediately on start up
sched/debug: Add deadline scheduler bandwidth ratio to /proc/sched_debug
sched/debug: Move sched_domain_sysctl to debug.c
sched/debug: Move the /sys/kernel/debug/sched_features file setup into debug.c
sched/rt: Fix PI handling vs. sched_setscheduler()
sched/core: Remove duplicated sched_group_set_shares() prototype
sched/fair: Consolidate nohz CPU load update code
sched/fair: Avoid using decay_load_missed() with a negative value
sched/deadline: Always calculate end of period on sched_yield()
sched/cgroup: Fix cgroup entity load tracking tear-down
rcu: Use simple wait queues where possible in rcutree
...
Functions which the compiler has instrumented for KASAN place poison on
the stack shadow upon entry and remove this poision prior to returning.
In the case of CPU hotplug, CPUs exit the kernel a number of levels deep
in C code. Any instrumented functions on this critical path will leave
portions of the stack shadow poisoned.
When a CPU is subsequently brought back into the kernel via a different
path, depending on stackframe, layout calls to instrumented functions
may hit this stale poison, resulting in (spurious) KASAN splats to the
console.
To avoid this, clear any stale poison from the idle thread for a CPU
prior to bringing a CPU online.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Reviewed-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Cc: Alexander Potapenko <glider@google.com>
Cc: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will.deacon@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The dl_new field of struct sched_dl_entity is currently used to
identify new deadline tasks, so that their deadline and runtime
can be properly initialised.
However, these tasks can be easily identified by checking if
their deadline is smaller than the current time when they switch
to SCHED_DEADLINE. So, dl_new can be removed by introducing this
check in switched_to_dl(); this allows to simplify the
SCHED_DEADLINE code.
Signed-off-by: Luca Abeni <luca.abeni@unitn.it>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Juri Lelli <juri.lelli@arm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1457350024-7825-2-git-send-email-luca.abeni@unitn.it
Signed-off-by: Ingo Molnar <mingo@kernel.org>
On CPU hotplug the steal time accounting can keep a stale rq->prev_steal_time
value over CPU down and up. So after the CPU comes up again the delta
calculation in steal_account_process_tick() wreckages itself due to the
unsigned math:
u64 steal = paravirt_steal_clock(smp_processor_id());
steal -= this_rq()->prev_steal_time;
So if steal is smaller than rq->prev_steal_time we end up with an insane large
value which then gets added to rq->prev_steal_time, resulting in a permanent
wreckage of the accounting. As a consequence the per CPU stats in /proc/stat
become stale.
Nice trick to tell the world how idle the system is (100%) while the CPU is
100% busy running tasks. Though we prefer realistic numbers.
None of the accounting values which use a previous value to account for
fractions is reset at CPU hotplug time. update_rq_clock_task() has a sanity
check for prev_irq_time and prev_steal_time_rq, but that sanity check solely
deals with clock warps and limits the /proc/stat visible wreckage. The
prev_time values are still wrong.
Solution is simple: Reset rq->prev_*_time when the CPU is plugged in again.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: <stable@vger.kernel.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Glauber Costa <glommer@parallels.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Fixes: commit 095c0aa83e "sched: adjust scheduler cpu power for stolen time"
Fixes: commit aa48380851 "sched: Remove irq time from available CPU power"
Fixes: commit e6e6685acc "KVM guest: Steal time accounting"
Link: http://lkml.kernel.org/r/alpine.DEB.2.11.1603041539490.3686@nanos
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Instead of providing asynchronous checks for the nohz subsystem to verify
sched tick dependency, migrate sched to the new mask.
Everytime a task is enqueued or dequeued, we evaluate the state of the
tick dependency on top of the policy of the tasks in the runqueue, by
order of priority:
SCHED_DEADLINE: Need the tick in order to periodically check for runtime
SCHED_FIFO : Don't need the tick (no round-robin)
SCHED_RR : Need the tick if more than 1 task of the same priority
for round robin (simplified with checking if more than
one SCHED_RR task no matter what priority).
SCHED_NORMAL : Need the tick if more than 1 task for round-robin.
We could optimize that further with one flag per sched policy on the tick
dependency mask and perform only the checks relevant to the policy
concerned by an enqueue/dequeue operation.
Since the checks aren't based on the current task anymore, we could get
rid of the task switch hook but it's still needed for posix cpu
timers.
Reviewed-by: Chris Metcalf <cmetcalf@ezchip.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Chris Metcalf <cmetcalf@ezchip.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Luiz Capitulino <lcapitulino@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Given that wq_worker_sleeping() could only be called for a
CPU it is running on, we do not need passing a CPU ID as an
argument.
Suggested-by: Oleg Nesterov <oleg@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Move the scheduler cpu online notifier part to the hotplug core. This is
anyway the highest priority callback and we need that functionality right now
for the next changes.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-arch@vger.kernel.org
Cc: Rik van Riel <riel@redhat.com>
Cc: Rafael Wysocki <rafael.j.wysocki@intel.com>
Cc: "Srivatsa S. Bhat" <srivatsa@mit.edu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Sebastian Siewior <bigeasy@linutronix.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Paul McKenney <paulmck@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul Turner <pjt@google.com>
Link: http://lkml.kernel.org/r/20160226182341.200791046@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The preempt_disable() invokes preempt_count_add() which saves the caller
in ->preempt_disable_ip. It uses CALLER_ADDR1 which does not look for
its caller but for the parent of the caller. Which means we get the correct
caller for something like spin_lock() unless the architectures inlines
those invocations. It is always wrong for preempt_disable() or
local_bh_disable().
This patch makes the function get_lock_parent_ip() which tries
CALLER_ADDR0,1,2 if the former is a locking function.
This seems to record the preempt_disable() caller properly for
preempt_disable() itself as well as for get_cpu_var() or
local_bh_disable().
Steven asked for the get_parent_ip() -> get_lock_parent_ip() rename.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20160226135456.GB18244@linutronix.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The sched_domain_sysctl setup is only enabled when SCHED_DEBUG is
configured. As debug.c is only compiled when SCHED_DEBUG is configured as
well, move the setup of sched_domain_sysctl into that file.
Note, the (un)register_sched_domain_sysctl() functions had to be changed
from static to allow access to them from core.c.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Clark Williams <williams@redhat.com>
Cc: Juri Lelli <juri.lelli@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20160222212825.599278093@goodmis.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
As /sys/kernel/debug/sched_features is only created when SCHED_DEBUG is enabled, and the file
debug.c is only compiled when SCHED_DEBUG is enabled, it makes sense to move
sched_feature setup into that file and get rid of the #ifdef.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Clark Williams <williams@redhat.com>
Cc: Juri Lelli <juri.lelli@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20160222212825.464193063@goodmis.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Andrea Parri reported:
> I found that the following scenario (with CONFIG_RT_GROUP_SCHED=y) is not
> handled correctly:
>
> T1 (prio = 20)
> lock(rtmutex);
>
> T2 (prio = 20)
> blocks on rtmutex (rt_nr_boosted = 0 on T1's rq)
>
> T1 (prio = 20)
> sys_set_scheduler(prio = 0)
> [new_effective_prio == oldprio]
> T1 prio = 20 (rt_nr_boosted = 0 on T1's rq)
>
> The last step is incorrect as T1 is now boosted (c.f., rt_se_boosted());
> in particular, if we continue with
>
> T1 (prio = 20)
> unlock(rtmutex)
> wakeup(T2)
> adjust_prio(T1)
> [prio != rt_mutex_getprio(T1)]
> dequeue(T1)
> rt_nr_boosted = (unsigned long)(-1)
> ...
> T1 prio = 0
>
> then we end up leaving rt_nr_boosted in an "inconsistent" state.
>
> The simple program attached could reproduce the previous scenario; note
> that, as a consequence of the presence of this state, the "assertion"
>
> WARN_ON(!rt_nr_running && rt_nr_boosted)
>
> from dec_rt_group() may trigger.
So normally we dequeue/enqueue tasks in sched_setscheduler(), which
would ensure the accounting stays correct. However in the early PI path
we fail to do so.
So this was introduced at around v3.14, by:
c365c292d0 ("sched: Consider pi boosting in setscheduler()")
which fixed another problem exactly because that dequeue/enqueue, joy.
Fix this by teaching rt about DEQUEUE_SAVE/ENQUEUE_RESTORE and have it
preserve runqueue location with that option. This requires decoupling
the on_rt_rq() state from being on the list.
In order to allow for explicit movement during the SAVE/RESTORE,
introduce {DE,EN}QUEUE_MOVE. We still must use SAVE/RESTORE in these
cases to preserve other invariants.
Respecting the SAVE/RESTORE flags also has the (nice) side-effect that
things like sys_nice()/sys_sched_setaffinity() also do not reorder
FIFO tasks (whereas they used to before this patch).
Reported-by: Andrea Parri <parri.andrea@gmail.com>
Tested-by: Andrea Parri <parri.andrea@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Juri Lelli <juri.lelli@arm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When a cgroup's CPU runqueue is destroyed, it should remove its
remaining load accounting from its parent cgroup.
The current site for doing so it unsuited because its far too late and
unordered against other cgroup removal (->css_free() will be, but we're also
in an RCU callback).
Put it in the ->css_offline() callback, which is the start of cgroup
destruction, right after the group has been made unavailable to
userspace. The ->css_offline() callbacks are called in hierarchical order
after the following v4.4 commit:
aa226ff4a1 ("cgroup: make sure a parent css isn't offlined before its children")
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Li Zefan <lizefan@huawei.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20160121212416.GL6357@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When CONFIG_GCOV is enabled, gcc decides to put context_switch()
out-of-line, which is inconsistent with its normal behavior.
It also causes an objtool warning because __schedule() no longer inlines
context_switch(), so the "STACK_FRAME_NON_STANDARD(__schedule)"
statement loses its effect.
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Bernd Petrovitsch <bernd@petrovitsch.priv.at>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Chris J Arges <chris.j.arges@canonical.com>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Michal Marek <mmarek@suse.cz>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Pedro Alves <palves@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: live-patching@vger.kernel.org
Link: http://lkml.kernel.org/r/d62aee926b6e303394e34a06999a964dc2773cf6.1456719558.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
objtool reports the following warnings for __schedule():
kernel/sched/core.o: warning: objtool:__schedule()+0x3c0: duplicate frame pointer save
kernel/sched/core.o: warning: objtool:__schedule()+0x3fd: sibling call from callable instruction with changed frame pointer
kernel/sched/core.o: warning: objtool:__schedule()+0x40a: call without frame pointer save/setup
kernel/sched/core.o: warning: objtool:__schedule()+0x7fd: frame pointer state mismatch
kernel/sched/core.o: warning: objtool:__schedule()+0x421: frame pointer state mismatch
Basically it's confused by two unusual attributes of the switch_to()
macro:
1. It saves prev's frame pointer to the old stack and restores next's
frame pointer from the new stack.
2. For new tasks it jumps directly to ret_from_fork.
Eventually it would probably be a good idea to clean up the
ret_from_fork hack so that new tasks are created with a valid initial
stack, as suggested by Andy:
https://lkml.kernel.org/r/CALCETrWsqCw4L1qKO9j9L5F+4ED4viuLQTFc=n1pKBZfFPQUFg@mail.gmail.com
Then __schedule() could return normally into the new code and objtool
hopefully wouldn't have a problem anymore.
In the meantime, mark its stack frame as non-standard so we can have a
baseline with no objtool warnings. The marker also serves as a reminder
that this code could be improved a bit.
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Bernd Petrovitsch <bernd@petrovitsch.priv.at>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Chris J Arges <chris.j.arges@canonical.com>
Cc: Jiri Slaby <jslaby@suse.cz>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Michal Marek <mmarek@suse.cz>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Pedro Alves <palves@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: live-patching@vger.kernel.org
Link: http://lkml.kernel.org/r/91190e324ebd7fcd01748d508d0dfd4693e84d91.1456719558.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Remove an unnecessary assignment of variable not used any more.
( This has no runtime effects as GCC is smart enough to optimize
this out. )
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1455159578-17256-1-git-send-email-byungchul.park@lge.com
[ Edited the changelog. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Export fetch_or() that's implemented and used internally by the
scheduler. We are going to use it for NO_HZ so make it generally
available.
Reviewed-by: Chris Metcalf <cmetcalf@ezchip.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Chris Metcalf <cmetcalf@ezchip.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Luiz Capitulino <lcapitulino@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
schedstats is very useful during debugging and performance tuning but it
incurs overhead to calculate the stats. As such, even though it can be
disabled at build time, it is often enabled as the information is useful.
This patch adds a kernel command-line and sysctl tunable to enable or
disable schedstats on demand (when it's built in). It is disabled
by default as someone who knows they need it can also learn to enable
it when necessary.
The benefits are dependent on how scheduler-intensive the workload is.
If it is then the patch reduces the number of cycles spent calculating
the stats with a small benefit from reducing the cache footprint of the
scheduler.
These measurements were taken from a 48-core 2-socket
machine with Xeon(R) E5-2670 v3 cpus although they were also tested on a
single socket machine 8-core machine with Intel i7-3770 processors.
netperf-tcp
4.5.0-rc1 4.5.0-rc1
vanilla nostats-v3r1
Hmean 64 560.45 ( 0.00%) 575.98 ( 2.77%)
Hmean 128 766.66 ( 0.00%) 795.79 ( 3.80%)
Hmean 256 950.51 ( 0.00%) 981.50 ( 3.26%)
Hmean 1024 1433.25 ( 0.00%) 1466.51 ( 2.32%)
Hmean 2048 2810.54 ( 0.00%) 2879.75 ( 2.46%)
Hmean 3312 4618.18 ( 0.00%) 4682.09 ( 1.38%)
Hmean 4096 5306.42 ( 0.00%) 5346.39 ( 0.75%)
Hmean 8192 10581.44 ( 0.00%) 10698.15 ( 1.10%)
Hmean 16384 18857.70 ( 0.00%) 18937.61 ( 0.42%)
Small gains here, UDP_STREAM showed nothing intresting and neither did
the TCP_RR tests. The gains on the 8-core machine were very similar.
tbench4
4.5.0-rc1 4.5.0-rc1
vanilla nostats-v3r1
Hmean mb/sec-1 500.85 ( 0.00%) 522.43 ( 4.31%)
Hmean mb/sec-2 984.66 ( 0.00%) 1018.19 ( 3.41%)
Hmean mb/sec-4 1827.91 ( 0.00%) 1847.78 ( 1.09%)
Hmean mb/sec-8 3561.36 ( 0.00%) 3611.28 ( 1.40%)
Hmean mb/sec-16 5824.52 ( 0.00%) 5929.03 ( 1.79%)
Hmean mb/sec-32 10943.10 ( 0.00%) 10802.83 ( -1.28%)
Hmean mb/sec-64 15950.81 ( 0.00%) 16211.31 ( 1.63%)
Hmean mb/sec-128 15302.17 ( 0.00%) 15445.11 ( 0.93%)
Hmean mb/sec-256 14866.18 ( 0.00%) 15088.73 ( 1.50%)
Hmean mb/sec-512 15223.31 ( 0.00%) 15373.69 ( 0.99%)
Hmean mb/sec-1024 14574.25 ( 0.00%) 14598.02 ( 0.16%)
Hmean mb/sec-2048 13569.02 ( 0.00%) 13733.86 ( 1.21%)
Hmean mb/sec-3072 12865.98 ( 0.00%) 13209.23 ( 2.67%)
Small gains of 2-4% at low thread counts and otherwise flat. The
gains on the 8-core machine were slightly different
tbench4 on 8-core i7-3770 single socket machine
Hmean mb/sec-1 442.59 ( 0.00%) 448.73 ( 1.39%)
Hmean mb/sec-2 796.68 ( 0.00%) 794.39 ( -0.29%)
Hmean mb/sec-4 1322.52 ( 0.00%) 1343.66 ( 1.60%)
Hmean mb/sec-8 2611.65 ( 0.00%) 2694.86 ( 3.19%)
Hmean mb/sec-16 2537.07 ( 0.00%) 2609.34 ( 2.85%)
Hmean mb/sec-32 2506.02 ( 0.00%) 2578.18 ( 2.88%)
Hmean mb/sec-64 2511.06 ( 0.00%) 2569.16 ( 2.31%)
Hmean mb/sec-128 2313.38 ( 0.00%) 2395.50 ( 3.55%)
Hmean mb/sec-256 2110.04 ( 0.00%) 2177.45 ( 3.19%)
Hmean mb/sec-512 2072.51 ( 0.00%) 2053.97 ( -0.89%)
In constract, this shows a relatively steady 2-3% gain at higher thread
counts. Due to the nature of the patch and the type of workload, it's
not a surprise that the result will depend on the CPU used.
hackbench-pipes
4.5.0-rc1 4.5.0-rc1
vanilla nostats-v3r1
Amean 1 0.0637 ( 0.00%) 0.0660 ( -3.59%)
Amean 4 0.1229 ( 0.00%) 0.1181 ( 3.84%)
Amean 7 0.1921 ( 0.00%) 0.1911 ( 0.52%)
Amean 12 0.3117 ( 0.00%) 0.2923 ( 6.23%)
Amean 21 0.4050 ( 0.00%) 0.3899 ( 3.74%)
Amean 30 0.4586 ( 0.00%) 0.4433 ( 3.33%)
Amean 48 0.5910 ( 0.00%) 0.5694 ( 3.65%)
Amean 79 0.8663 ( 0.00%) 0.8626 ( 0.43%)
Amean 110 1.1543 ( 0.00%) 1.1517 ( 0.22%)
Amean 141 1.4457 ( 0.00%) 1.4290 ( 1.16%)
Amean 172 1.7090 ( 0.00%) 1.6924 ( 0.97%)
Amean 192 1.9126 ( 0.00%) 1.9089 ( 0.19%)
Some small gains and losses and while the variance data is not included,
it's close to the noise. The UMA machine did not show anything particularly
different
pipetest
4.5.0-rc1 4.5.0-rc1
vanilla nostats-v2r2
Min Time 4.13 ( 0.00%) 3.99 ( 3.39%)
1st-qrtle Time 4.38 ( 0.00%) 4.27 ( 2.51%)
2nd-qrtle Time 4.46 ( 0.00%) 4.39 ( 1.57%)
3rd-qrtle Time 4.56 ( 0.00%) 4.51 ( 1.10%)
Max-90% Time 4.67 ( 0.00%) 4.60 ( 1.50%)
Max-93% Time 4.71 ( 0.00%) 4.65 ( 1.27%)
Max-95% Time 4.74 ( 0.00%) 4.71 ( 0.63%)
Max-99% Time 4.88 ( 0.00%) 4.79 ( 1.84%)
Max Time 4.93 ( 0.00%) 4.83 ( 2.03%)
Mean Time 4.48 ( 0.00%) 4.39 ( 1.91%)
Best99%Mean Time 4.47 ( 0.00%) 4.39 ( 1.91%)
Best95%Mean Time 4.46 ( 0.00%) 4.38 ( 1.93%)
Best90%Mean Time 4.45 ( 0.00%) 4.36 ( 1.98%)
Best50%Mean Time 4.36 ( 0.00%) 4.25 ( 2.49%)
Best10%Mean Time 4.23 ( 0.00%) 4.10 ( 3.13%)
Best5%Mean Time 4.19 ( 0.00%) 4.06 ( 3.20%)
Best1%Mean Time 4.13 ( 0.00%) 4.00 ( 3.39%)
Small improvement and similar gains were seen on the UMA machine.
The gain is small but it stands to reason that doing less work in the
scheduler is a good thing. The downside is that the lack of schedstats and
tracepoints may be surprising to experts doing performance analysis until
they find the existence of the schedstats= parameter or schedstats sysctl.
It will be automatically activated for latencytop and sleep profiling to
alleviate the problem. For tracepoints, there is a simple warning as it's
not safe to activate schedstats in the context when it's known the tracepoint
may be wanted but is unavailable.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Reviewed-by: Matt Fleming <matt@codeblueprint.co.uk>
Reviewed-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <mgalbraith@suse.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1454663316-22048-1-git-send-email-mgorman@techsingularity.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The isolcpus= kernel boot parameter restricts userspace from scheduling on
the specified CPUs.
If a CPU is specified that is outside the range of 0 to nr_cpu_ids,
cpulist_parse() will return -ERANGE, return an empty cpulist, and
fail silently.
This patch adds an error message to isolated_cpu_setup() to indicate to
the user that something has gone awry, and returns 0 on error.
Signed-off-by: Prarit Bhargava <prarit@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1454596680-10367-1-git-send-email-prarit@redhat.com
[ Twiddled some details. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull scheduler fixes from Thomas Gleixner:
"Three small fixes in the scheduler/core:
- use after free in the numa code
- crash in the numa init code
- a simple spelling fix"
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
pid: Fix spelling in comments
sched/numa: Fix use-after-free bug in the task_numa_compare
sched: Fix crash in sched_init_numa()
parallel to mutex_{lock,unlock,trylock,is_locked,lock_nested},
inode_foo(inode) being mutex_foo(&inode->i_mutex).
Please, use those for access to ->i_mutex; over the coming cycle
->i_mutex will become rwsem, with ->lookup() done with it held
only shared.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
The following PowerPC commit:
c118baf802 ("arch/powerpc/mm/numa.c: do not allocate bootmem memory for non existing nodes")
avoids allocating bootmem memory for non existent nodes.
But when DEBUG_PER_CPU_MAPS=y is enabled, my powerNV system failed to boot
because in sched_init_numa(), cpumask_or() operation was done on
unallocated nodes.
Fix that by making cpumask_or() operation only on existing nodes.
[ Tested with and w/o DEBUG_PER_CPU_MAPS=y on x86 and PowerPC. ]
Reported-by: Jan Stancek <jstancek@redhat.com>
Tested-by: Jan Stancek <jstancek@redhat.com>
Signed-off-by: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com>
Cc: <gkurz@linux.vnet.ibm.com>
Cc: <grant.likely@linaro.org>
Cc: <nikunj@linux.vnet.ibm.com>
Cc: <vdavydov@parallels.com>
Cc: <linuxppc-dev@lists.ozlabs.org>
Cc: <linux-mm@kvack.org>
Cc: <peterz@infradead.org>
Cc: <benh@kernel.crashing.org>
Cc: <paulus@samba.org>
Cc: <mpe@ellerman.id.au>
Cc: <anton@samba.org>
Link: http://lkml.kernel.org/r/1452884483-11676-1-git-send-email-raghavendra.kt@linux.vnet.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull cgroup updates from Tejun Heo:
- cgroup v2 interface is now official. It's no longer hidden behind a
devel flag and can be mounted using the new cgroup2 fs type.
Unfortunately, cpu v2 interface hasn't made it yet due to the
discussion around in-process hierarchical resource distribution and
only memory and io controllers can be used on the v2 interface at the
moment.
- The existing documentation which has always been a bit of mess is
relocated under Documentation/cgroup-v1/. Documentation/cgroup-v2.txt
is added as the authoritative documentation for the v2 interface.
- Some features are added through for-4.5-ancestor-test branch to
enable netfilter xt_cgroup match to use cgroup v2 paths. The actual
netfilter changes will be merged through the net tree which pulled in
the said branch.
- Various cleanups
* 'for-4.5' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
cgroup: rename cgroup documentations
cgroup: fix a typo.
cgroup: Remove resource_counter.txt in Documentation/cgroup-legacy/00-INDEX.
cgroup: demote subsystem init messages to KERN_DEBUG
cgroup: Fix uninitialized variable warning
cgroup: put controller Kconfig options in meaningful order
cgroup: clean up the kernel configuration menu nomenclature
cgroup_pids: fix a typo.
Subject: cgroup: Fix incomplete dd command in blkio documentation
cgroup: kill cgrp_ss_priv[CGROUP_CANFORK_COUNT] and friends
cpuset: Replace all instances of time_t with time64_t
cgroup: replace unified-hierarchy.txt with a proper cgroup v2 documentation
cgroup: rename Documentation/cgroups/ to Documentation/cgroup-legacy/
cgroup: replace __DEVEL__sane_behavior with cgroup2 fs type
Pull scheduler updates from Ingo Molnar:
"The main changes in this cycle were:
- tickless load average calculation enhancements (Byungchul Park)
- vtime handling enhancements (Frederic Weisbecker)
- scalability improvement via properly aligning a key structure field
(Jiri Olsa)
- various stop_machine() fixes (Oleg Nesterov)
- sched/numa enhancement (Rik van Riel)
- various fixes and improvements (Andi Kleen, Dietmar Eggemann,
Geliang Tang, Hiroshi Shimamoto, Joonwoo Park, Peter Zijlstra,
Waiman Long, Wanpeng Li, Yuyang Du)"
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (32 commits)
sched/fair: Fix new task's load avg removed from source CPU in wake_up_new_task()
sched/core: Move sched_entity::avg into separate cache line
x86/fpu: Properly align size in CHECK_MEMBER_AT_END_OF() macro
sched/deadline: Fix the earliest_dl.next logic
sched/fair: Disable the task group load_avg update for the root_task_group
sched/fair: Move the cache-hot 'load_avg' variable into its own cacheline
sched/fair: Avoid redundant idle_cpu() call in update_sg_lb_stats()
sched/core: Move the sched_to_prio[] arrays out of line
sched/cputime: Convert vtime_seqlock to seqcount
sched/cputime: Introduce vtime accounting check for readers
sched/cputime: Rename vtime_accounting_enabled() to vtime_accounting_cpu_enabled()
sched/cputime: Correctly handle task guest time on housekeepers
sched/cputime: Clarify vtime symbols and document them
sched/cputime: Remove extra cost in task_cputime()
sched/fair: Make it possible to account fair load avg consistently
sched/fair: Modify the comment about lock assumptions in migrate_task_rq_fair()
stop_machine: Clean up the usage of the preemption counter in cpu_stopper_thread()
stop_machine: Shift the 'done != NULL' check from cpu_stop_signal_done() to callers
stop_machine: Kill cpu_stop_done->executed
stop_machine: Change __stop_cpus() to rely on cpu_stop_queue_work()
...
Pull locking updates from Ingo Molnar:
"So we have a laundry list of locking subsystem changes:
- continuing barrier API and code improvements
- futex enhancements
- atomics API improvements
- pvqspinlock enhancements: in particular lock stealing and adaptive
spinning
- qspinlock micro-enhancements"
* 'locking-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
futex: Allow FUTEX_CLOCK_REALTIME with FUTEX_WAIT op
futex: Cleanup the goto confusion in requeue_pi()
futex: Remove pointless put_pi_state calls in requeue()
futex: Document pi_state refcounting in requeue code
futex: Rename free_pi_state() to put_pi_state()
futex: Drop refcount if requeue_pi() acquired the rtmutex
locking/barriers, arch: Remove ambiguous statement in the smp_store_mb() documentation
lcoking/barriers, arch: Use smp barriers in smp_store_release()
locking/cmpxchg, arch: Remove tas() definitions
locking/pvqspinlock: Queue node adaptive spinning
locking/pvqspinlock: Allow limited lock stealing
locking/pvqspinlock: Collect slowpath lock statistics
sched/core, locking: Document Program-Order guarantees
locking, sched: Introduce smp_cond_acquire() and use it
locking/pvqspinlock, x86: Optimize the PV unlock code path
locking/qspinlock: Avoid redundant read of next pointer
locking/qspinlock: Prefetch the next node cacheline
locking/qspinlock: Use _acquire/_release() versions of cmpxchg() & xchg()
atomics: Add test for atomic operations with _relaxed variants
Pull RCU changes from Paul E. McKenney:
- Adding transitivity uniformly to rcu_node structure ->lock
acquisitions. (This is implemented by the first two commits
on top of v4.4-rc2 due to the pervasive nature of this change.)
- Documentation updates, including RCU requirements.
- Expedited grace-period changes.
- Miscellaneous fixes.
- Linked-list fixes, courtesy of KTSAN.
- Torture-test updates.
- Late-breaking fix to sysrq-generated crash.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The following commit which went into mainline through networking tree
3b13758f51 ("cgroups: Allow dynamically changing net_classid")
conflicts in net/core/netclassid_cgroup.c with the following pending
fix in cgroup/for-4.4-fixes.
1f7dd3e5a6 ("cgroup: fix handling of multi-destination migration from subtree_control enabling")
The former separates out update_classid() from cgrp_attach() and
updates it to walk all fds of all tasks in the target css so that it
can be used from both migration and config change paths. The latter
drops @css from cgrp_attach().
Resolve the conflict by making cgrp_attach() call update_classid()
with the css from the first task. We can revive @tset walking in
cgrp_attach() but given that net_cls is v1 only where there always is
only one target css during migration, this is fine.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Nina Schiff <ninasc@fb.com>
We need the scheduler's fastpaths to be, well, fast, and unnecessarily
disabling and re-enabling interrupts is not necessarily consistent with
this goal. Especially given that there are regions of the scheduler that
already have interrupts disabled.
This commit therefore moves the call to rcu_note_context_switch()
to one of the interrupts-disabled regions of the scheduler, and
removes the now-redundant disabling and re-enabling of interrupts from
rcu_note_context_switch() and the functions it calls.
Reported-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
[ paulmck: Shift rcu_note_context_switch() to avoid deadlock, as suggested
by Peter Zijlstra. ]
If a system with large number of sockets was driven to full
utilization, it was found that the clock tick handling occupied a
rather significant proportion of CPU time when fair group scheduling
and autogroup were enabled.
Running a java benchmark on a 16-socket IvyBridge-EX system, the perf
profile looked like:
10.52% 0.00% java [kernel.vmlinux] [k] smp_apic_timer_interrupt
9.66% 0.05% java [kernel.vmlinux] [k] hrtimer_interrupt
8.65% 0.03% java [kernel.vmlinux] [k] tick_sched_timer
8.56% 0.00% java [kernel.vmlinux] [k] update_process_times
8.07% 0.03% java [kernel.vmlinux] [k] scheduler_tick
6.91% 1.78% java [kernel.vmlinux] [k] task_tick_fair
5.24% 5.04% java [kernel.vmlinux] [k] update_cfs_shares
In particular, the high CPU time consumed by update_cfs_shares()
was mostly due to contention on the cacheline that contained the
task_group's load_avg statistical counter. This cacheline may also
contains variables like shares, cfs_rq & se which are accessed rather
frequently during clock tick processing.
This patch moves the load_avg variable into another cacheline
separated from the other frequently accessed variables. It also
creates a cacheline aligned kmemcache for task_group to make sure
that all the allocated task_group's are cacheline aligned.
By doing so, the perf profile became:
9.44% 0.00% java [kernel.vmlinux] [k] smp_apic_timer_interrupt
8.74% 0.01% java [kernel.vmlinux] [k] hrtimer_interrupt
7.83% 0.03% java [kernel.vmlinux] [k] tick_sched_timer
7.74% 0.00% java [kernel.vmlinux] [k] update_process_times
7.27% 0.03% java [kernel.vmlinux] [k] scheduler_tick
5.94% 1.74% java [kernel.vmlinux] [k] task_tick_fair
4.15% 3.92% java [kernel.vmlinux] [k] update_cfs_shares
The %cpu time is still pretty high, but it is better than before. The
benchmark results before and after the patch was as follows:
Before patch - Max-jOPs: 907533 Critical-jOps: 134877
After patch - Max-jOPs: 916011 Critical-jOps: 142366
Signed-off-by: Waiman Long <Waiman.Long@hpe.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Ben Segall <bsegall@google.com>
Cc: Douglas Hatch <doug.hatch@hpe.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: Paul Turner <pjt@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Scott J Norton <scott.norton@hpe.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Yuyang Du <yuyang.du@intel.com>
Link: http://lkml.kernel.org/r/1449081710-20185-3-git-send-email-Waiman.Long@hpe.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When building a kernel with a gcc 6 snapshot the compiler complains
about unused const static variables for prio_to_weight and prio_to_mult
for multiple scheduler files (all but core.c and autogroup.c)
The way the array is currently declared it will be duplicated in
every scheduler file that includes sched.h, which seems rather wasteful.
Move the array out of line into core.c. I also added a sched_ prefix
to avoid any potential name space collisions.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1448859583-3252-1-git-send-email-andi@firstfloor.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The current code accounts for the time a task was absent from the fair
class (per ATTACH_AGE_LOAD). However it does not work correctly when a
task got migrated or moved to another cgroup while outside of the fair
class.
This patch tries to address that by aging on migration. We locklessly
read the 'last_update_time' stamp from both the old and new cfs_rq,
ages the load upto the old time, and sets it to the new time.
These timestamps should in general not be more than 1 tick apart from
one another, so there is a definite bound on things.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
[ Changelog, a few edits and !SMP build fix ]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1445616981-29904-2-git-send-email-byungchul.park@lge.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
These are some notes on the scheduler locking and how it provides
program order guarantees on SMP systems.
( This commit is in the locking tree, because the new documentation
refers to a newly introduced locking primitive. )
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will.deacon@arm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Introduce smp_cond_acquire() which combines a control dependency and a
read barrier to form acquire semantics.
This primitive has two benefits:
- it documents control dependencies,
- its typically cheaper than using smp_load_acquire() in a loop.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Oleg noticed that its possible to falsely observe p->on_cpu == 0 such
that we'll prematurely continue with the wakeup and effectively run p on
two CPUs at the same time.
Even though the overlap is very limited; the task is in the middle of
being scheduled out; it could still result in corruption of the
scheduler data structures.
CPU0 CPU1
set_current_state(...)
<preempt_schedule>
context_switch(X, Y)
prepare_lock_switch(Y)
Y->on_cpu = 1;
finish_lock_switch(X)
store_release(X->on_cpu, 0);
try_to_wake_up(X)
LOCK(p->pi_lock);
t = X->on_cpu; // 0
context_switch(Y, X)
prepare_lock_switch(X)
X->on_cpu = 1;
finish_lock_switch(Y)
store_release(Y->on_cpu, 0);
</preempt_schedule>
schedule();
deactivate_task(X);
X->on_rq = 0;
if (X->on_rq) // false
if (t) while (X->on_cpu)
cpu_relax();
context_switch(X, ..)
finish_lock_switch(X)
store_release(X->on_cpu, 0);
Avoid the load of X->on_cpu being hoisted over the X->on_rq load.
Reported-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Explain how the control dependency and smp_rmb() end up providing
ACQUIRE semantics and pair with smp_store_release() in
finish_lock_switch().
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
root_domain::rto_mask allocated through alloc_cpumask_var()
contains garbage data, this may cause problems. For instance,
When doing pull_rt_task(), it may do useless iterations if
rto_mask retains some extra garbage bits. Worse still, this
violates the isolated domain rule for clustered scheduling
using cpuset, because the tasks(with all the cpus allowed)
belongs to one root domain can be pulled away into another
root domain.
The patch cleans the garbage by using zalloc_cpumask_var()
instead of alloc_cpumask_var() for root_domain::rto_mask
allocation, thereby addressing the issues.
Do the same thing for root_domain's other cpumask memembers:
dlo_mask, span, and online.
Signed-off-by: Xunlei Pang <xlpang@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: <stable@vger.kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1449057179-29321-1-git-send-email-xlpang@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Because wakeups can (fundamentally) be late, a task might not be in
the expected state. Therefore testing against a task's state is racy,
and can yield false positives.
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: oleg@redhat.com
Fixes: 9067ac85d5 ("wake_up_process() should be never used to wakeup a TASK_STOPPED/TRACED task")
Link: http://lkml.kernel.org/r/1448933660-23082-1-git-send-email-sasha.levin@oracle.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Now that nobody use the "priv" arg passed to can_fork/cancel_fork/fork we can
kill CGROUP_CANFORK_COUNT/SUBSYS_TAG/etc and cgrp_ss_priv[] in copy_process().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Consider the following v2 hierarchy.
P0 (+memory) --- P1 (-memory) --- A
\- B
P0 has memory enabled in its subtree_control while P1 doesn't. If
both A and B contain processes, they would belong to the memory css of
P1. Now if memory is enabled on P1's subtree_control, memory csses
should be created on both A and B and A's processes should be moved to
the former and B's processes the latter. IOW, enabling controllers
can cause atomic migrations into different csses.
The core cgroup migration logic has been updated accordingly but the
controller migration methods haven't and still assume that all tasks
migrate to a single target css; furthermore, the methods were fed the
css in which subtree_control was updated which is the parent of the
target csses. pids controller depends on the migration methods to
move charges and this made the controller attribute charges to the
wrong csses often triggering the following warning by driving a
counter negative.
WARNING: CPU: 1 PID: 1 at kernel/cgroup_pids.c:97 pids_cancel.constprop.6+0x31/0x40()
Modules linked in:
CPU: 1 PID: 1 Comm: systemd Not tainted 4.4.0-rc1+ #29
...
ffffffff81f65382 ffff88007c043b90 ffffffff81551ffc 0000000000000000
ffff88007c043bc8 ffffffff810de202 ffff88007a752000 ffff88007a29ab00
ffff88007c043c80 ffff88007a1d8400 0000000000000001 ffff88007c043bd8
Call Trace:
[<ffffffff81551ffc>] dump_stack+0x4e/0x82
[<ffffffff810de202>] warn_slowpath_common+0x82/0xc0
[<ffffffff810de2fa>] warn_slowpath_null+0x1a/0x20
[<ffffffff8118e031>] pids_cancel.constprop.6+0x31/0x40
[<ffffffff8118e0fd>] pids_can_attach+0x6d/0xf0
[<ffffffff81188a4c>] cgroup_taskset_migrate+0x6c/0x330
[<ffffffff81188e05>] cgroup_migrate+0xf5/0x190
[<ffffffff81189016>] cgroup_attach_task+0x176/0x200
[<ffffffff8118949d>] __cgroup_procs_write+0x2ad/0x460
[<ffffffff81189684>] cgroup_procs_write+0x14/0x20
[<ffffffff811854e5>] cgroup_file_write+0x35/0x1c0
[<ffffffff812e26f1>] kernfs_fop_write+0x141/0x190
[<ffffffff81265f88>] __vfs_write+0x28/0xe0
[<ffffffff812666fc>] vfs_write+0xac/0x1a0
[<ffffffff81267019>] SyS_write+0x49/0xb0
[<ffffffff81bcef32>] entry_SYSCALL_64_fastpath+0x12/0x76
This patch fixes the bug by removing @css parameter from the three
migration methods, ->can_attach, ->cancel_attach() and ->attach() and
updating cgroup_taskset iteration helpers also return the destination
css in addition to the task being migrated. All controllers are
updated accordingly.
* Controllers which don't care whether there are one or multiple
target csses can be converted trivially. cpu, io, freezer, perf,
netclassid and netprio fall in this category.
* cpuset's current implementation assumes that there's single source
and destination and thus doesn't support v2 hierarchy already. The
only change made by this patchset is how that single destination css
is obtained.
* memory migration path already doesn't do anything on v2. How the
single destination css is obtained is updated and the prep stage of
mem_cgroup_can_attach() is reordered to accomodate the change.
* pids is the only controller which was affected by this bug. It now
correctly handles multi-destination migrations and no longer causes
counter underflow from incorrect accounting.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-and-tested-by: Daniel Wagner <daniel.wagner@bmw-carit.de>
Cc: Aleksa Sarai <cyphar@cyphar.com>
